revil-O/platypush
1
0
Fork 0
forked from platypush/platypush
Code Pull requests Activity

Sound plugin rewritten almost from scratch.

Browse source
This commit is contained in:
Fabio Manganiello 2023-06-27 13:31:38 +02:00
parent 2fb6e4d7d0
commit f2540437b7
Signed by untrusted user: blacklight
GPG key ID: D90FBA7F76362774
34 changed files with 2931 additions and 1586 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

55
platypush/message/event/sound.py
View file

@ -1,65 +1,70 @@
from abc import ABC
from typing import Optional, Tuple, Union
from platypush.message.event import Event
class SoundEvent(Event):
class SoundEvent(Event, ABC):
"""Base class for sound events"""
def __init__(self, *args, **kwargs):
super().__init__(*args, **kwargs)
def __init__(
self, *args, device: Optional[Union[str, Tuple[str, str]]] = None, **kwargs
):
super().__init__(*args, device=device, **kwargs)
class SoundPlaybackStartedEvent(SoundEvent):
class SoundEventWithResource(SoundEvent, ABC):
"""Base class for sound events with resource names attached"""
def __init__(self, *args, resource: Optional[str] = None, **kwargs):
super().__init__(*args, resource=resource, **kwargs)
class SoundPlaybackStartedEvent(SoundEventWithResource):
"""
Event triggered when a new sound playback starts
"""
def __init__(self, filename=None, *args, **kwargs):
super().__init__(*args, filename=filename, **kwargs)
class SoundPlaybackStoppedEvent(SoundEvent):
class SoundPlaybackStoppedEvent(SoundEventWithResource):
"""
Event triggered when the sound playback stops
"""
def __init__(self, filename=None, *args, **kwargs):
super().__init__(*args, filename=filename, **kwargs)
class SoundPlaybackPausedEvent(SoundEvent):
class SoundPlaybackPausedEvent(SoundEventWithResource):
"""
Event triggered when the sound playback pauses
"""
def __init__(self, *args, **kwargs):
super().__init__(*args, **kwargs)
class SoundPlaybackResumedEvent(SoundEventWithResource):
"""
Event triggered when the sound playback resumsed from a paused state
"""
class SoundRecordingStartedEvent(SoundEvent):
class SoundRecordingStartedEvent(SoundEventWithResource):
"""
Event triggered when a new recording starts
"""
def __init__(self, filename=None, *args, **kwargs):
super().__init__(*args, filename=filename, **kwargs)
class SoundRecordingStoppedEvent(SoundEvent):
class SoundRecordingStoppedEvent(SoundEventWithResource):
"""
Event triggered when a sound recording stops
"""
def __init__(self, filename=None, *args, **kwargs):
super().__init__(*args, filename=filename, **kwargs)
class SoundRecordingPausedEvent(SoundEvent):
class SoundRecordingPausedEvent(SoundEventWithResource):
"""
Event triggered when a sound recording pauses
"""
def __init__(self, *args, **kwargs):
super().__init__(*args, **kwargs)
class SoundRecordingResumedEvent(SoundEvent):
"""
Event triggered when a sound recording resumes from a paused state
"""
# vim:sw=4:ts=4:et:

911
platypush/plugins/sound/__init__.py
View file

File diff suppressed because it is too large Load diff

5
platypush/plugins/sound/_controllers/__init__.py
View file

@ -1,5 +0,0 @@
from ._base import AudioThread
from ._recorder import AudioRecorder
__all__ = ['AudioRecorder', 'AudioThread']

227
platypush/plugins/sound/_controllers/_base.py
View file

@ -1,227 +0,0 @@
from contextlib import contextmanager
import queue
import time
from abc import ABC, abstractmethod
from logging import getLogger
from threading import Event, RLock, Thread
from typing import IO, Generator, Optional, Tuple, Union
from typing_extensions import override
import sounddevice as sd
from .._converter import ConverterProcess
from .._model import AudioState
class AudioThread(Thread, ABC):
"""
Base class for audio play/record threads.
"""
_STREAM_NAME_PREFIX = 'platypush-stream-'
def __init__(
self,
plugin,
device: Union[str, Tuple[str, str]],
outfile: str,
output_format: str,
channels: int,
sample_rate: int,
dtype: str,
stream: bool,
audio_pass_through: bool,
duration: Optional[float] = None,
blocksize: Optional[int] = None,
latency: Union[float, str] = 'high',
redis_queue: Optional[str] = None,
should_stop: Optional[Event] = None,
**kwargs,
):
from .. import SoundPlugin
super().__init__(**kwargs)
self.plugin: SoundPlugin = plugin
self.device = device
self.outfile = outfile
self.output_format = output_format
self.channels = channels
self.sample_rate = sample_rate
self.dtype = dtype
self.stream = stream
self.duration = duration
self.blocksize = blocksize
self.latency = latency
self.redis_queue = redis_queue
self.audio_pass_through = audio_pass_through
self.logger = getLogger(__name__)
self._state = AudioState.STOPPED
self._state_lock = RLock()
self._started_time: Optional[float] = None
self._converter: Optional[ConverterProcess] = None
self._should_stop = should_stop or Event()
self.paused_changed = Event()
@property
def should_stop(self) -> bool:
"""
Proxy for `._should_stop.is_set()`.
"""
return self._should_stop.is_set()
@abstractmethod
def _audio_callback(self, audio_converter: ConverterProcess):
"""
Returns a callback to handle the raw frames captures from the audio device.
"""
raise NotImplementedError()
@abstractmethod
def _on_audio_converted(self, data: bytes, out_f: IO):
"""
This callback will be called when the audio data has been converted.
"""
raise NotImplementedError()
def main(
self,
converter: ConverterProcess,
audio_stream: sd.Stream,
out_stream_index: Optional[int],
out_f: IO,
):
"""
Main loop.
"""
self.notify_start()
if out_stream_index:
self.plugin.start_playback(
stream_index=out_stream_index, stream=audio_stream
)
self.logger.info(
'Started %s on device [%s]', self.__class__.__name__, self.device
)
self._started_time = time.time()
while (
self.state != AudioState.STOPPED
and not self.should_stop
and (
self.duration is None
or time.time() - self._started_time < self.duration
)
):
while self.state == AudioState.PAUSED:
self.paused_changed.wait()
if self.should_stop:
break
timeout = (
max(
0,
self.duration - (time.time() - self._started_time),
)
if self.duration is not None
else 1
)
data = converter.read(timeout=timeout)
if not data:
continue
self._on_audio_converted(data, out_f)
@override
def run(self):
super().run()
self.paused_changed.clear()
try:
stream_index = (
self.plugin._allocate_stream_index() # pylint: disable=protected-access
if self.audio_pass_through
else None
)
with self.open_converter() as converter, sd.Stream(
samplerate=self.sample_rate,
device=self.device,
channels=self.channels,
callback=self._audio_callback(converter),
dtype=self.dtype,
latency=self.latency,
blocksize=self.blocksize,
) as audio_stream, open(self.outfile, 'wb') as f:
self.main(
out_stream_index=stream_index,
converter=converter,
audio_stream=audio_stream,
out_f=f,
)
except queue.Empty:
self.logger.warning(
'Audio callback timeout for %s', self.__class__.__name__
)
finally:
self.notify_stop()
@contextmanager
def open_converter(self) -> Generator[ConverterProcess, None, None]:
assert not self._converter, 'A converter process is already running'
self._converter = ConverterProcess(
ffmpeg_bin=self.plugin.ffmpeg_bin,
sample_rate=self.sample_rate,
channels=self.channels,
dtype=self.dtype,
chunk_size=self.plugin.input_blocksize,
output_format=self.output_format,
)
self._converter.start()
yield self._converter
self._converter.stop()
self._converter.join(timeout=2)
self._converter = None
def notify_start(self):
self.state = AudioState.RUNNING
def notify_stop(self):
self.state = AudioState.STOPPED
if self._converter:
self._converter.stop()
def notify_pause(self):
states = {
AudioState.PAUSED: AudioState.RUNNING,
AudioState.RUNNING: AudioState.PAUSED,
}
with self._state_lock:
new_state = states.get(self.state)
if new_state:
self.state = new_state
else:
return
self.logger.info('Paused state toggled for %s', self.__class__.__name__)
self.paused_changed.set()
@property
def state(self):
with self._state_lock:
return self._state
@state.setter
def state(self, value: AudioState):
with self._state_lock:
self._state = value
# vim:sw=4:ts=4:et:

69
platypush/plugins/sound/_controllers/_recorder.py
View file

@ -1,69 +0,0 @@
from typing import IO
from typing_extensions import override
from platypush.context import get_bus
from platypush.message.event.sound import (
SoundRecordingStartedEvent,
SoundRecordingStoppedEvent,
)
from platypush.utils import get_redis
from .._converter import ConverterProcess
from .._model import AudioState
from ._base import AudioThread
class AudioRecorder(AudioThread):
"""
The ``AudioRecorder`` thread is responsible for recording audio from the
input device, writing it to the converter process and dispatch the
converted audio to the registered consumers.
"""
def __init__(self, *args, **kwargs):
super().__init__(*args, **kwargs)
@override
def _audio_callback(self, audio_converter: ConverterProcess):
# _ = frames
# __ = time
def callback(indata, outdata, _, __, status):
if self.state != AudioState.RUNNING:
return
if status:
self.logger.warning('Recording callback status: %s', status)
try:
audio_converter.write(indata.tobytes())
except AssertionError as e:
self.logger.warning('Audio converter callback error: %s', e)
self.state = AudioState.STOPPED
return
if self.audio_pass_through:
outdata[:] = indata
return callback
@override
def _on_audio_converted(self, data: bytes, out_f: IO):
out_f.write(data)
if self.redis_queue and self.stream:
get_redis().publish(self.redis_queue, data)
@override
def notify_start(self):
super().notify_start()
get_bus().post(SoundRecordingStartedEvent())
@override
def notify_stop(self):
prev_state = self.state
super().notify_stop()
if prev_state != AudioState.STOPPED:
get_bus().post(SoundRecordingStoppedEvent())
# vim:sw=4:ts=4:et:

183
platypush/plugins/sound/_converter.py
View file

@ -1,183 +0,0 @@
import asyncio
from asyncio.subprocess import PIPE
from logging import getLogger
from queue import Empty
from queue import Queue
from threading import Event, RLock, Thread
from typing import Optional, Self
from platypush.context import get_or_create_event_loop
_dtype_to_ffmpeg_format = {
'int8': 's8',
'uint8': 'u8',
'int16': 's16le',
'uint16': 'u16le',
'int32': 's32le',
'uint32': 'u32le',
'float32': 'f32le',
'float64': 'f64le',
}
"""
Supported input types:
'int8', 'uint8', 'int16', 'uint16', 'int32', 'uint32', 'float32', 'float64'
"""
_output_format_to_ffmpeg_args = {
'wav': ('-f', 'wav'),
'ogg': ('-f', 'ogg'),
'mp3': ('-f', 'mp3'),
'aac': ('-f', 'adts'),
'flac': ('-f', 'flac'),
}
class ConverterProcess(Thread):
"""
Wrapper for an ffmpeg converter instance.
"""
def __init__(
self,
ffmpeg_bin: str,
sample_rate: int,
channels: int,
dtype: str,
chunk_size: int,
output_format: str,
*args,
**kwargs,
):
"""
:param ffmpeg_bin: Path to the ffmpeg binary.
:param sample_rate: The sample rate of the input audio.
:param channels: The number of channels of the input audio.
:param dtype: The (numpy) data type of the raw input audio.
:param chunk_size: Number of bytes that will be read at once from the
ffmpeg process.
:param output_format: Output audio format.
"""
super().__init__(*args, **kwargs)
ffmpeg_format = _dtype_to_ffmpeg_format.get(dtype)
assert ffmpeg_format, (
f'Unsupported data type: {dtype}. Supported data types: '
f'{list(_dtype_to_ffmpeg_format.keys())}'
)
self._ffmpeg_bin = ffmpeg_bin
self._ffmpeg_format = ffmpeg_format
self._sample_rate = sample_rate
self._channels = channels
self._chunk_size = chunk_size
self._output_format = output_format
self._closed = False
self._out_queue = Queue()
self.ffmpeg = None
self.logger = getLogger(__name__)
self._loop = None
self._should_stop = Event()
self._stop_lock = RLock()
def __enter__(self) -> Self:
self.start()
return self
def __exit__(self, *_, **__):
self.stop()
def _check_ffmpeg(self):
assert (
self.ffmpeg and self.ffmpeg.returncode is None
), 'The ffmpeg process has already terminated'
def _get_format_args(self):
ffmpeg_args = _output_format_to_ffmpeg_args.get(self._output_format)
assert ffmpeg_args, (
f'Unsupported output format: {self._output_format}. Supported formats: '
f'{list(_output_format_to_ffmpeg_args.keys())}'
)
return ffmpeg_args
async def _audio_proxy(self, timeout: Optional[float] = None):
self.ffmpeg = await asyncio.create_subprocess_exec(
self._ffmpeg_bin,
'-f',
self._ffmpeg_format,
'-ar',
str(self._sample_rate),
'-ac',
str(self._channels),
'-i',
'pipe:',
*self._get_format_args(),
'pipe:',
stdin=PIPE,
stdout=PIPE,
)
try:
await asyncio.wait_for(self.ffmpeg.wait(), 0.1)
except asyncio.TimeoutError:
pass
while (
self._loop
and self.ffmpeg
and self.ffmpeg.returncode is None
and not self.should_stop
):
self._check_ffmpeg()
assert (
self.ffmpeg and self.ffmpeg.stdout
), 'The stdout is closed for the ffmpeg process'
try:
data = await asyncio.wait_for(
self.ffmpeg.stdout.read(self._chunk_size), timeout
)
self._out_queue.put(data)
except asyncio.TimeoutError:
self._out_queue.put(b'')
def write(self, data: bytes):
self._check_ffmpeg()
assert (
self.ffmpeg and self._loop and self.ffmpeg.stdin
), 'The stdin is closed for the ffmpeg process'
self._loop.call_soon_threadsafe(self.ffmpeg.stdin.write, data)
def read(self, timeout: Optional[float] = None) -> Optional[bytes]:
try:
return self._out_queue.get(timeout=timeout)
except Empty:
return None
def run(self):
super().run()
self._loop = get_or_create_event_loop()
try:
self._loop.run_until_complete(self._audio_proxy(timeout=1))
except RuntimeError as e:
self.logger.warning(e)
finally:
self.stop()
def stop(self):
with self._stop_lock:
self._should_stop.set()
if self.ffmpeg:
self.ffmpeg.kill()
self.ffmpeg = None
self._loop = None
@property
def should_stop(self) -> bool:
return self._should_stop.is_set()
# vim:sw=4:ts=4:et:

10
platypush/plugins/sound/_converters/__init__.py Normal file
View file

@ -0,0 +1,10 @@
from ._base import AudioConverter
from ._from_raw import RawInputAudioConverter
from ._to_raw import RawOutputAudioConverter, RawOutputAudioFromFileConverter
__all__ = [
'AudioConverter',
'RawInputAudioConverter',
'RawOutputAudioConverter',
'RawOutputAudioFromFileConverter',
]

331
platypush/plugins/sound/_converters/_base.py Normal file
View file

@ -0,0 +1,331 @@
from abc import ABC, abstractmethod
import asyncio
from asyncio.subprocess import PIPE
from logging import getLogger
from queue import Empty, Queue
from threading import Event, RLock, Thread
from typing import Any, Callable, Coroutine, Iterable, Optional, Self
from platypush.context import get_or_create_event_loop
_dtype_to_ffmpeg_format = {
'int8': 's8',
'uint8': 'u8',
'int16': 's16le',
'uint16': 'u16le',
'int32': 's32le',
'uint32': 'u32le',
'float32': 'f32le',
'float64': 'f64le',
}
"""
Supported raw types:
'int8', 'uint8', 'int16', 'uint16', 'int32', 'uint32', 'float32', 'float64'
"""
class AudioConverter(Thread, ABC):
"""
Base class for an ffmpeg audio converter instance.
"""
_format_to_ffmpeg_args = {
'wav': ('-f', 'wav'),
'ogg': ('-f', 'ogg'),
'mp3': ('-f', 'mp3'),
'aac': ('-f', 'adts'),
'flac': ('-f', 'flac'),
}
def __init__(
self,
*args,
ffmpeg_bin: str,
sample_rate: int,
channels: int,
volume: float,
dtype: str,
chunk_size: int,
format: Optional[str] = None, # pylint: disable=redefined-builtin
on_exit: Optional[Callable[[], Any]] = None,
**kwargs,
):
"""
:param ffmpeg_bin: Path to the ffmpeg binary.
:param sample_rate: The sample rate of the input/output audio.
:param channels: The number of channels of the input/output audio.
:param volume: Audio volume, as a percentage between 0 and 100.
:param dtype: The (numpy) data type of the raw input/output audio.
:param chunk_size: Number of bytes that will be read at once from the
ffmpeg process.
:param format: Input/output audio format.
:param on_exit: Function to call when the ffmpeg process exits.
"""
super().__init__(*args, **kwargs)
ffmpeg_format = _dtype_to_ffmpeg_format.get(dtype)
assert ffmpeg_format, (
f'Unsupported data type: {dtype}. Supported data types: '
f'{list(_dtype_to_ffmpeg_format.keys())}'
)
self._ffmpeg_bin = ffmpeg_bin
self._ffmpeg_format = ffmpeg_format
self._ffmpeg_task: Optional[Coroutine] = None
self._sample_rate = sample_rate
self._channels = channels
self._chunk_size = chunk_size
self._format = format
self._closed = False
self._out_queue = Queue()
self.ffmpeg = None
self.volume = volume
self.logger = getLogger(__name__)
self._loop = None
self._should_stop = Event()
self._stop_lock = RLock()
self._on_exit = on_exit
self._ffmpeg_terminated = Event()
def __enter__(self) -> Self:
"""
Audio converter context manager.
It starts and registers the ffmpeg converter process.
"""
self.start()
return self
def __exit__(self, *_, **__):
"""
Audio converter context manager.
It stops and unregisters the ffmpeg converter process.
"""
self.stop()
def _check_ffmpeg(self):
assert not self.terminated, 'The ffmpeg process has already terminated'
@property
def gain(self) -> float:
return self.volume / 100
@property
def terminated(self) -> bool:
return self._ffmpeg_terminated.is_set()
@property
def _default_args(self) -> Iterable[str]:
"""
Set of arguments common to all ffmpeg converter instances.
"""
return ('-hide_banner', '-loglevel', 'warning', '-y')
@property
@abstractmethod
def _input_format_args(self) -> Iterable[str]:
"""
Ffmpeg audio input arguments.
"""
raise NotImplementedError()
@property
@abstractmethod
def _output_format_args(self):
"""
Ffmpeg audio output arguments.
"""
raise NotImplementedError()
@property
def _channel_layout_args(self) -> Iterable[str]:
"""
Set of extra ffmpeg arguments for the channel layout.
"""
args = ('-ac', str(self._channels))
if self._channels == 1:
return args + ('-channel_layout', 'mono')
if self._channels == 2:
return args + ('-channel_layout', 'stereo')
return args
@property
def _raw_ffmpeg_args(self) -> Iterable[str]:
"""
Ffmpeg arguments for raw audio input/output given the current
configuration.
"""
return (
'-f',
self._ffmpeg_format,
'-ar',
str(self._sample_rate),
*self._channel_layout_args,
)
@property
def _audio_volume_args(self) -> Iterable[str]:
"""
Ffmpeg audio volume arguments.
"""
return ('-filter:a', f'volume={self.gain}')
@property
def _input_source_args(self) -> Iterable[str]:
"""
Default arguments for the ffmpeg input source (default: ``-i pipe:``,
ffmpeg will read from a pipe filled by the application).
"""
return ('-i', 'pipe:')
@property
def _output_target_args(self) -> Iterable[str]:
"""
Default arguments for the ffmpeg output target (default: ``pipe:``,
ffmpeg will write the output to a pipe read by the application).
"""
return ('pipe:',)
@property
def _converter_stdin(self) -> Optional[int]:
"""
Default stdin file descriptor to be used by the ffmpeg converter.
Default: ``PIPE``, as the ffmpeg process by default reads audio frames
from the stdin.
"""
return PIPE
@property
def _compressed_ffmpeg_args(self) -> Iterable[str]:
"""
Ffmpeg arguments for the compressed audio given the current
configuration.
"""
if not self._format:
return ()
ffmpeg_args = self._format_to_ffmpeg_args.get(self._format)
assert ffmpeg_args, (
f'Unsupported output format: {self._format}. Supported formats: '
f'{list(self._format_to_ffmpeg_args.keys())}'
)
return ffmpeg_args
async def _audio_proxy(self, timeout: Optional[float] = None):
"""
Proxy the converted audio stream to the output queue for downstream
consumption.
"""
ffmpeg_args = (
self._ffmpeg_bin,
*self._default_args,
*self._input_format_args,
*self._input_source_args,
*self._output_format_args,
*self._output_target_args,
)
self.ffmpeg = await asyncio.create_subprocess_exec(
*ffmpeg_args,
stdin=self._converter_stdin,
stdout=PIPE,
)
self.logger.info('Running ffmpeg: %s', ' '.join(ffmpeg_args))
try:
await asyncio.wait_for(self.ffmpeg.wait(), 0.1)
except asyncio.TimeoutError:
pass
while (
self._loop
and self.ffmpeg
and self.ffmpeg.returncode is None
and not self.should_stop
):
self._check_ffmpeg()
assert (
self.ffmpeg and self.ffmpeg.stdout
), 'The stdout is closed for the ffmpeg process'
self._ffmpeg_terminated.clear()
try:
data = await asyncio.wait_for(
self.ffmpeg.stdout.read(self._chunk_size), timeout
)
self._out_queue.put(data)
except asyncio.TimeoutError:
self._out_queue.put(b'')
def write(self, data: bytes):
"""
Write raw data to the ffmpeg process.
"""
self._check_ffmpeg()
assert (
self.ffmpeg and self._loop and self.ffmpeg.stdin
), 'The stdin is closed for the ffmpeg process'
self._loop.call_soon_threadsafe(self.ffmpeg.stdin.write, data)
def read(self, timeout: Optional[float] = None) -> Optional[bytes]:
"""
Read the next chunk of converted audio bytes from the converter queue.
"""
try:
return self._out_queue.get(timeout=timeout)
except Empty:
return None
def run(self):
"""
Main runner. It runs the audio proxy in a loop and cleans up everything
in case of stop/failure.
"""
super().run()
self._loop = get_or_create_event_loop()
try:
self._ffmpeg_task = self._audio_proxy(timeout=1)
self._loop.run_until_complete(self._ffmpeg_task)
except RuntimeError as e:
self.logger.warning(e)
finally:
self.stop()
def stop(self):
"""
Sets the stop event, kills the ffmpeg process and resets the context.
"""
with self._stop_lock:
self._should_stop.set()
if self._ffmpeg_task:
self._ffmpeg_task.close()
self._ffmpeg_task = None
try:
if self.ffmpeg and self.ffmpeg.returncode is None:
self.ffmpeg.kill()
except ProcessLookupError:
pass
self.ffmpeg = None
self._loop = None
self._ffmpeg_terminated.set()
if self._on_exit:
self._on_exit()
@property
def should_stop(self) -> bool:
"""
Proxy property for the ``_should_stop`` event.
"""
return self._should_stop.is_set()
# vim:sw=4:ts=4:et:

23
platypush/plugins/sound/_converters/_from_raw.py Normal file
View file

@ -0,0 +1,23 @@
from typing import Iterable
from typing_extensions import override
from ._base import AudioConverter
class RawInputAudioConverter(AudioConverter):
"""
Converts raw audio input to a compressed media format.
"""
@property
@override
def _input_format_args(self) -> Iterable[str]:
return self._raw_ffmpeg_args
@property
@override
def _output_format_args(self) -> Iterable[str]:
return self._compressed_ffmpeg_args
# vim:sw=4:ts=4:et:

38
platypush/plugins/sound/_converters/_to_raw.py Normal file
View file

@ -0,0 +1,38 @@
from typing import Iterable
from typing_extensions import override
from ._base import AudioConverter
class RawOutputAudioConverter(AudioConverter):
"""
Converts input audio to raw audio output.
"""
@property
@override
def _input_format_args(self) -> Iterable[str]:
return self._compressed_ffmpeg_args
@property
@override
def _output_format_args(self) -> Iterable[str]:
return self._raw_ffmpeg_args
class RawOutputAudioFromFileConverter(RawOutputAudioConverter):
"""
Converts an input file to raw audio output.
"""
def __init__(self, *args, infile: str, **kwargs):
super().__init__(*args, **kwargs)
self.infile = infile
@property
@override
def _input_source_args(self) -> Iterable[str]:
return ('-i', self.infile)
# vim:sw=4:ts=4:et:

3
platypush/plugins/sound/_manager/__init__.py Normal file
View file

@ -0,0 +1,3 @@
from ._main import AudioManager
__all__ = ["AudioManager"]

91
platypush/plugins/sound/_manager/_device.py Normal file
View file

@ -0,0 +1,91 @@
from typing import List, Optional
import sounddevice as sd
from .._model import AudioDevice, DeviceType, StreamType
class DeviceManager:
"""
The device manager is responsible for managing the virtual audio device
abstractions exposed by the OS.
For example, on a pure ALSA system virtual devices are usually mapped the
physical audio devices available on the system.
On a system that runs through PulseAudio or Jack, there may be a
``default`` virtual device whose sound card mappings may be managed by the
audio server.
"""
def __init__(
self,
input_device: Optional[DeviceType] = None,
output_device: Optional[DeviceType] = None,
):
"""
:param input_device: The default input device to use (by index or name).
:param output_device: The default output device to use (by index or name).
"""
self.input_device = (
self.get_device(input_device, StreamType.INPUT)
if input_device is not None
else None
)
self.output_device = (
self.get_device(output_device, StreamType.OUTPUT)
if output_device is not None
else None
)
def get_devices(
self, type: Optional[StreamType] = None # pylint: disable=redefined-builtin
) -> List[AudioDevice]:
"""
Get available audio devices.
:param type: The type of devices to filter (default: return all).
"""
devices: List[dict] = sd.query_devices() # type: ignore
if type:
devices = [dev for dev in devices if dev.get(f'max_{type.value}_channels')]
return [AudioDevice(**info) for info in devices]
def get_device(
self,
device: Optional[DeviceType] = None,
type: Optional[StreamType] = None, # pylint: disable=redefined-builtin
) -> AudioDevice:
"""
Search for a device.
Either ``device`` or ``type`` have to be specified.
:param device: The device to search for, either by index or name. If
not specified, then the default device for the given type is
returned.
:param type: The type of the device to search.
"""
assert device or type, 'Please specify either device or type'
if device is None:
if type == StreamType.INPUT and self.input_device is not None:
return self.input_device
if type == StreamType.OUTPUT and self.output_device is not None:
return self.output_device
try:
info: dict = sd.query_devices(
kind=type.value if type else None, device=device # type: ignore
)
except sd.PortAudioError as e:
raise AssertionError(
f'Could not get device for type={type} and device={device}: {e}',
type,
device,
e,
) from e
assert info, f'No such device: {device}'
return AudioDevice(**info)

291
platypush/plugins/sound/_manager/_main.py Normal file
View file

@ -0,0 +1,291 @@
from logging import getLogger
import os
import stat
from threading import Event
from time import time
from typing import Iterable, List, Optional, Union
from .._model import AudioDevice, DeviceType, StreamType
from .._streams import AudioPlayer, AudioRecorder, AudioThread
from ._device import DeviceManager
from ._stream import StreamManager
class AudioManager:
"""
The audio manager is responsible for managing multiple audio controllers and
their access to audio resources.
It main purpose is to act as a proxy/facade between the high-level audio
plugin and the audio functionalities (allocating streams, managing the state
of the player and recorder processes, etc.).
"""
_default_signal_timeout = 2
def __init__(
self,
should_stop: Event,
input_blocksize: int,
output_blocksize: int,
input_device: Optional[DeviceType] = None,
output_device: Optional[DeviceType] = None,
queue_size: Optional[int] = None,
):
"""
:param should_stop: Event to synchronize the audio manager stop.
:param input_blocksize: Block size for the input stream.
:param output_blocksize: Block size for the output stream.
:param input_device: Default device to use for the input stream.
:param output_device: Default device to use for the output stream.
:param queue_size: Maximum size of the audio queues.
"""
self._should_stop = should_stop
self._device_manager = DeviceManager(
input_device=input_device, output_device=output_device
)
self._stream_manager = StreamManager(device_manager=self._device_manager)
self.logger = getLogger(__name__)
self.input_blocksize = input_blocksize
self.output_blocksize = output_blocksize
self.queue_size = queue_size
def create_player(
self,
device: DeviceType,
channels: int,
volume: float,
infile: Optional[str] = None,
sound: Optional[Union[dict, Iterable[dict]]] = None,
duration: Optional[float] = None,
sample_rate: Optional[int] = None,
dtype: str = 'int16',
blocksize: Optional[int] = None,
latency: Union[float, str] = 'high',
stream_name: Optional[str] = None,
) -> AudioPlayer:
"""
Create an audio player thread.
:param device: Audio device to use.
:param channels: Number of output channels.
:param volume: Output volume, between 0 and 100.
:param infile: File or URL to play.
:param sound: Alternatively to a file/URL, you can play synthetic
sounds.
:param duration: Duration of the stream in seconds.
:param sample_rate: Sample rate of the stream.
:param dtype: Data type of the stream.
:param blocksize: Block size of the stream.
:param latency: Latency of the stream.
:param stream_name: Name of the stream.
"""
dev = self._device_manager.get_device(device, type=StreamType.OUTPUT)
player = AudioPlayer.build(
device=device,
infile=infile,
sound=sound,
duration=duration,
volume=volume,
sample_rate=sample_rate or dev.default_samplerate,
dtype=dtype,
blocksize=blocksize or self.output_blocksize,
latency=latency,
channels=channels,
queue_size=self.queue_size,
should_stop=self._should_stop,
)
self._stream_manager.register(
player, dev, StreamType.OUTPUT, stream_name=stream_name
)
return player
def create_recorder(
self,
device: DeviceType,
output_device: Optional[DeviceType] = None,
fifo: Optional[str] = None,
outfile: Optional[str] = None,
duration: Optional[float] = None,
sample_rate: Optional[int] = None,
dtype: str = 'int16',
blocksize: Optional[int] = None,
latency: Union[float, str] = 'high',
channels: int = 1,
volume: float = 100,
redis_queue: Optional[str] = None,
format: str = 'wav', # pylint: disable=redefined-builtin
stream: bool = True,
stream_name: Optional[str] = None,
play_audio: bool = False,
) -> AudioRecorder:
"""
Create an audio recorder thread.
:param device: Audio device to use.
:param output_device: Output device to use.
:param fifo: Path to an output FIFO file to use to synchronize the audio
to other processes.
:param outfile: Optional output file for the recorded audio.
:param duration: Duration of the recording in seconds.
:param sample_rate: Sample rate of the stream.
:param dtype: Data type of the stream.
:param blocksize: Block size of the stream.
:param latency: Latency of the stream.
:param channels: Number of output channels.
:param volume: Input volume, between 0 and 100.
:param redis_queue: Name of the Redis queue to use.
:param format: Format of the recorded audio.
:param stream: Whether to stream the recorded audio.
:param play_audio: Whether to play the recorded audio in real-time.
:param stream_name: Name of the stream.
"""
blocksize = blocksize or self.input_blocksize
dev = self._device_manager.get_device(device, type=StreamType.OUTPUT)
if fifo:
fifo = os.path.expanduser(fifo)
if os.path.exists(fifo) and stat.S_ISFIFO(os.stat(fifo).st_mode):
self.logger.info('Removing previous input stream FIFO %s', fifo)
os.unlink(fifo)
os.mkfifo(fifo, 0o644)
outfile = fifo
elif outfile:
outfile = os.path.expanduser(outfile)
outfile = outfile or fifo or os.devnull
recorder = AudioRecorder(
device=(
(
dev.index,
self._device_manager.get_device(
type=StreamType.OUTPUT, device=output_device
).index,
)
if play_audio
else dev.index
),
outfile=outfile,
duration=duration,
sample_rate=sample_rate or dev.default_samplerate,
dtype=dtype,
blocksize=blocksize,
latency=latency,
output_format=format,
channels=channels,
volume=volume,
redis_queue=redis_queue,
stream=stream,
audio_pass_through=play_audio,
queue_size=self.queue_size,
should_stop=self._should_stop,
)
self._stream_manager.register(
recorder, dev, StreamType.INPUT, stream_name=stream_name
)
return recorder
def get_device(
self,
device: Optional[DeviceType] = None,
type: Optional[StreamType] = None, # pylint: disable=redefined-builtin
) -> AudioDevice:
"""
Proxy to ``self._device_manager.get_device``.
"""
return self._device_manager.get_device(device=device, type=type)
def get_devices(
self,
type: Optional[StreamType] = None, # pylint: disable=redefined-builtin
) -> List[AudioDevice]:
"""
Proxy to ``self._device_manager.get_devices``.
"""
return self._device_manager.get_devices(type=type)
def get_streams(
self,
device: Optional[DeviceType] = None,
type: Optional[StreamType] = None, # pylint: disable=redefined-builtin
streams: Optional[Iterable[Union[str, int]]] = None,
) -> List[AudioThread]:
"""
Proxy to ``self._stream_manager.get``.
"""
return self._stream_manager.get(device=device, type=type, streams=streams)
def stop_audio(
self,
device: Optional[DeviceType] = None,
type: Optional[StreamType] = None, # pylint: disable=redefined-builtin
streams: Optional[Iterable[Union[str, int]]] = None,
timeout: Optional[float] = 2,
):
"""
Stops audio sessions.
:param device: Filter by host audio device.
:param type: Filter by stream type (input or output).
:param streams: Filter by stream indices/names.
:param timeout: Wait timeout in seconds.
"""
streams_to_stop = self._stream_manager.get(device, type, streams=streams)
# Send the stop signals
for audio_thread in streams_to_stop:
audio_thread.notify_stop()
# Wait for termination (with timeout)
wait_start = time()
for audio_thread in streams_to_stop:
audio_thread.join(
timeout=max(0, timeout - (time() - wait_start))
if timeout is not None
else None
)
# Remove references
for audio_thread in streams_to_stop:
self._stream_manager.unregister(audio_thread)
def pause_audio(
self,
device: Optional[DeviceType] = None,
type: Optional[StreamType] = None, # pylint: disable=redefined-builtin
streams: Optional[Iterable[Union[str, int]]] = None,
):
"""
Pauses/resumes audio sessions.
:param device: Filter by host audio device.
:param type: Filter by stream type (input or output).
:param streams: Filter by stream indices/names.
"""
streams_to_pause = self._stream_manager.get(device, type, streams=streams)
# Send the pause toggle signals
for audio_thread in streams_to_pause:
audio_thread.notify_pause()
def set_volume(
self,
volume: float,
device: Optional[DeviceType] = None,
streams: Optional[Iterable[Union[str, int]]] = None,
):
"""
:param volume: New volume, between 0 and 100.
:param device: Set the volume only on the specified device (default:
all).
:param streams: Set the volume only on the specified list of stream
indices/names (default: all).
"""
stream_objs = self._stream_manager.get(device=device, streams=streams)
for stream in stream_objs:
stream.volume = volume

207
platypush/plugins/sound/_manager/_stream.py Normal file
View file

@ -0,0 +1,207 @@
from collections import defaultdict
from logging import getLogger
from threading import RLock
from typing import Dict, Iterable, List, Optional, Union
from .._model import AudioDevice, DeviceType, StreamType
from .._streams import AudioThread
from ._device import DeviceManager
class StreamManager:
"""
The audio manager is responsible for storing the current state of the
playing/recording audio streams and allowing fast flexible lookups (by
stream index, name, type, device, and any combination of those).
"""
def __init__(self, device_manager: DeviceManager):
"""
:param device_manager: Reference to the device manager.
"""
self._next_stream_index = 1
self._device_manager = device_manager
self._state_lock = RLock()
self._stream_index_by_name: Dict[str, int] = {}
self._stream_name_by_index: Dict[int, str] = {}
self._stream_index_to_device: Dict[int, AudioDevice] = {}
self._stream_index_to_type: Dict[int, StreamType] = {}
self.logger = getLogger(__name__)
self._streams: Dict[
int, Dict[StreamType, Dict[int, AudioThread]]
] = defaultdict(lambda: {stream_type: {} for stream_type in StreamType})
""" {device_index: {stream_type: {stream_index: audio_thread}}} """
self._streams_by_index: Dict[StreamType, Dict[int, AudioThread]] = {
stream_type: {} for stream_type in StreamType
}
""" {stream_type: {stream_index: [audio_threads]}} """
self._stream_locks: Dict[int, Dict[StreamType, RLock]] = defaultdict(
lambda: {stream_type: RLock() for stream_type in StreamType}
)
""" {device_index: {stream_type: RLock}} """
@classmethod
def _generate_stream_name(
cls,
type: StreamType, # pylint: disable=redefined-builtin
stream_index: int,
) -> str:
return f'platypush:audio:{type.value}:{stream_index}'
def _gen_next_stream_index(
self,
type: StreamType, # pylint: disable=redefined-builtin
stream_name: Optional[str] = None,
) -> int:
"""
:param type: The type of the stream to allocate (input or output).
:param stream_name: The name of the stream to allocate.
:return: The index of the new stream.
"""
with self._state_lock:
stream_index = self._next_stream_index
if not stream_name:
stream_name = self._generate_stream_name(type, stream_index)
self._stream_name_by_index[stream_index] = stream_name
self._stream_index_by_name[stream_name] = stream_index
self._next_stream_index += 1
return stream_index
def register(
self,
audio_thread: AudioThread,
device: AudioDevice,
type: StreamType, # pylint: disable=redefined-builtin
stream_name: Optional[str] = None,
):
"""
Registers an audio stream to a device.
:param audio_thread: Stream to register.
:param device: Device to register the stream to.
:param type: The type of the stream to allocate (input or output).
:param stream_name: The name of the stream to allocate.
"""
with self._state_lock:
stream_index = audio_thread.stream_index
if stream_index is None:
stream_index = audio_thread.stream_index = self._gen_next_stream_index(
type, stream_name=stream_name
)
self._streams[device.index][type][stream_index] = audio_thread
self._stream_index_to_device[stream_index] = device
self._stream_index_to_type[stream_index] = type
self._streams_by_index[type][stream_index] = audio_thread
def unregister(
self,
audio_thread: AudioThread,
device: Optional[AudioDevice] = None,
type: Optional[StreamType] = None, # pylint: disable=redefined-builtin
):
"""
Unregisters an audio stream from a device.
:param audio_thread: Stream to unregister.
:param device: Device to unregister the stream from.
:param type: The type of the stream to unregister (input or output).
"""
with self._state_lock:
stream_index = audio_thread.stream_index
if stream_index is None:
return
if device is None:
device = self._stream_index_to_device.get(stream_index)
if not type:
type = self._stream_index_to_type.get(stream_index)
if device is None or type is None:
return
self._streams[device.index][type].pop(stream_index, None)
self._stream_index_to_device.pop(stream_index, None)
self._stream_index_to_type.pop(stream_index, None)
self._streams_by_index[type].pop(stream_index, None)
stream_name = self._stream_name_by_index.pop(stream_index, None)
if stream_name:
self._stream_index_by_name.pop(stream_name, None)
def _get_by_device_and_type(
self,
device: Optional[DeviceType] = None,
type: Optional[StreamType] = None, # pylint: disable=redefined-builtin
) -> List[AudioThread]:
"""
Filter streams by device and/or type.
"""
devs = (
[self._device_manager.get_device(device, type)]
if device is not None
else self._device_manager.get_devices(type)
)
return [
audio_thread
for dev in devs
for stream_info in (
[self._streams[dev.index].get(type, {})]
if type
else list(self._streams[dev.index].values())
)
for audio_thread in stream_info.values()
if audio_thread and audio_thread.is_alive()
]
def _get_by_stream_index_or_name(
self, streams: Iterable[Union[str, int]]
) -> List[AudioThread]:
"""
Filter streams by index or name.
"""
threads = []
for stream in streams:
try:
stream_index = int(stream)
except (TypeError, ValueError):
stream_index = self._stream_index_by_name.get(stream) # type: ignore
if stream_index is None:
self.logger.warning('No such audio stream: %s', stream)
continue
stream_type = self._stream_index_to_type.get(stream_index)
if not stream_type:
self.logger.warning(
'No type available for this audio stream: %s', stream
)
continue
thread = self._streams_by_index.get(stream_type, {}).get(stream_index)
if thread:
threads.append(thread)
return threads
def get(
self,
device: Optional[DeviceType] = None,
type: Optional[StreamType] = None, # pylint: disable=redefined-builtin
streams: Optional[Iterable[Union[str, int]]] = None,
) -> List[AudioThread]:
"""
Searches streams, either by device and/or type, or by stream index/name.
"""
return (
self._get_by_stream_index_or_name(streams)
if streams
else self._get_by_device_and_type(device, type)
)

31
platypush/plugins/sound/_model.py
View file

@ -1,4 +1,26 @@
from dataclasses import dataclass
from enum import Enum
from typing import Union
DeviceType = Union[int, str]
@dataclass
class AudioDevice:
"""
Maps the properties of an audio device.
"""
index: int
name: str
hostapi: int
max_input_channels: int
max_output_channels: int
default_samplerate: int
default_low_input_latency: float = 0
default_low_output_latency: float = 0
default_high_input_latency: float = 0
default_high_output_latency: float = 0
class AudioState(Enum):
@ -9,3 +31,12 @@ class AudioState(Enum):
STOPPED = 'STOPPED'
RUNNING = 'RUNNING'
PAUSED = 'PAUSED'
class StreamType(Enum):
"""
Stream types.
"""
INPUT = 'input'
OUTPUT = 'output'

6
platypush/plugins/sound/_streams/__init__.py Normal file
View file

@ -0,0 +1,6 @@
from ._base import AudioThread
from ._player import AudioPlayer
from ._recorder import AudioRecorder
__all__ = ['AudioPlayer', 'AudioRecorder', 'AudioThread']

502
platypush/plugins/sound/_streams/_base.py Normal file
View file

@ -0,0 +1,502 @@
from abc import ABC, abstractmethod
from contextlib import contextmanager
from datetime import datetime
from logging import getLogger
import os
import queue
from threading import Event, RLock, Thread
import time
from typing import IO, Callable, Final, Generator, Optional, Tuple, Type, Union
from typing_extensions import override
import sounddevice as sd
from platypush.context import get_bus
from platypush.message.event.sound import SoundEvent
from platypush.utils import get_redis
from .._converters import AudioConverter
from .._model import AudioState, StreamType
_StreamType = Union[sd.Stream, sd.OutputStream]
class AudioThread(Thread, ABC):
"""
Base class for audio play/record stream threads.
"""
_DEFAULT_FILE: Final[str] = os.devnull
"""Unless otherwise specified, the audio streams will be sent to /dev/null"""
_DEFAULT_CONVERTER_TIMEOUT: Final[float] = 1
def __init__(
self,
device: Union[str, Tuple[str, str]],
channels: int,
volume: float,
sample_rate: int,
dtype: str,
blocksize: int,
ffmpeg_bin: str = 'ffmpeg',
stream: bool = False,
audio_pass_through: bool = False,
infile: Optional[str] = None,
outfile: Optional[str] = None,
duration: Optional[float] = None,
latency: Union[float, str] = 'high',
redis_queue: Optional[str] = None,
should_stop: Optional[Event] = None,
converter_timeout: Optional[float] = None,
stream_name: Optional[str] = None,
queue_size: Optional[int] = None,
**kwargs,
):
"""
:param device: Audio device to use.
:param channels: Number of channels to use.
:param volume: Input/output volume, between 0 and 100.
:param sample_rate: Sample rate to use.
:param dtype: Data type to use.
:param blocksize: Block size to use.
:param ffmpeg_bin: Path to the ffmpeg binary.
:param stream: Whether to stream the audio to Redis consumers.
:param audio_pass_through: Whether to pass the audio through to the
application's output stream.
:param infile: Path to the input file or URL, if this is an output
stream.
:param outfile: Path to the output file.
:param duration: Duration of the audio stream.
:param latency: Latency to use.
:param redis_queue: Redis queue to use.
:param should_stop: Synchronize with upstream stop events.
:param converter_timeout: How long to wait for the converter to finish.
:param stream_name: Name of the stream.
:param queue_size: Maximum size of the audio queue.
"""
super().__init__(**kwargs)
self.device = device
self.outfile = os.path.expanduser(outfile or self._DEFAULT_FILE)
self.infile = os.path.expanduser(infile or self._DEFAULT_FILE)
self.ffmpeg_bin = ffmpeg_bin
self.channels = channels
self.volume = volume
self.sample_rate = sample_rate
self.dtype = dtype
self.stream = stream
self.duration = duration
self.blocksize = blocksize * channels
self.latency = latency
self._redis_queue = redis_queue
self.audio_pass_through = audio_pass_through
self.queue_size = queue_size
self._stream_name = stream_name
self.logger = getLogger(__name__)
self._state = AudioState.STOPPED
self._state_lock = RLock()
self._started_time: Optional[float] = None
self._converter: Optional[AudioConverter] = None
self._should_stop = should_stop or Event()
self._converter_timeout = converter_timeout or self._DEFAULT_CONVERTER_TIMEOUT
self.audio_stream: Optional[_StreamType] = None
self.stream_index: Optional[int] = None
self.paused_changed = Event()
self._converter_terminated = Event()
@property
def should_stop(self) -> bool:
"""
Proxy for `._should_stop.is_set()`.
"""
return self._should_stop.is_set() or bool(
self.state == AudioState.STOPPED and self._started_time
)
@property
def gain(self) -> float:
return self.volume / 100
def wait_stop(self, timeout: Optional[float] = None):
"""
Wait for the stop signal to be received.
"""
return self._should_stop.wait(timeout=timeout)
def _audio_callback(self) -> Callable:
"""
Returns a callback to handle the raw frames captures from the audio device.
"""
def empty_callback(*_, **__):
pass
return empty_callback
@property
def stream_name(self) -> str:
if self._stream_name:
return self._stream_name
ret = f'platypush:audio:{self.direction.value}'
if self.stream_index is not None:
ret += f':{self.stream_index}'
return ret
@stream_name.setter
def stream_name(self, value: Optional[str]):
self._stream_name = value
@property
@abstractmethod
def direction(self) -> StreamType:
"""
The default direction for this stream - input or output.
"""
raise NotImplementedError()
@property
@abstractmethod
def _audio_converter_type(self) -> Optional[Type[AudioConverter]]:
"""
This property indicates the type that should be used for the audio
converter.
"""
raise NotImplementedError()
@property
@abstractmethod
def _started_event_type(self) -> Type[SoundEvent]:
"""
Event type that will be emitted when the audio starts.
"""
raise NotImplementedError()
@property
@abstractmethod
def _stopped_event_type(self) -> Type[SoundEvent]:
"""
Event type that will be emitted when the audio stops.
"""
raise NotImplementedError()
@property
@abstractmethod
def _paused_event_type(self) -> Type[SoundEvent]:
"""
Event type that will be emitted when the audio is paused.
"""
raise NotImplementedError()
@property
@abstractmethod
def _resumed_event_type(self) -> Type[SoundEvent]:
"""
Event type that will be emitted when the audio is resumed.
"""
raise NotImplementedError()
@property
def _stream_type(self) -> Union[Type[sd.Stream], Type[sd.OutputStream]]:
"""
The type of stream this thread is mapped to.
"""
return sd.Stream
@property
def _converter_args(self) -> dict:
"""
Extra arguments to pass to the audio converter.
"""
return {}
@property
def _stream_args(self) -> dict:
"""
Extra arguments to pass to the stream constructor.
"""
return {}
@property
def redis_queue(self) -> str:
"""
Redis queue for audio streaming.
"""
if self._redis_queue:
return self._redis_queue
dev = (
self.device
if isinstance(self.device, (str, int))
else '-'.join(map(str, self.device))
)
name = f'platypush-audio-stream-{self.__class__.__name__}-{dev}'
if self.stream_index is not None:
name = f'{name}-{self.stream_index}'
return name
def _on_audio_converted(self, data: bytes, out_f: Optional[IO] = None):
"""
This callback will be called when the audio data has been converted.
"""
if out_f:
out_f.write(data)
if self.stream:
get_redis().publish(self.redis_queue, data)
def _wait_running(self):
"""
If the stream is in paused state, wait for the state to change.
"""
while self.state == AudioState.PAUSED:
self.paused_changed.wait()
def main(
self,
converter: Optional[AudioConverter] = None,
out_f: Optional[IO] = None,
):
"""
Main loop.
"""
self.notify_start()
self.logger.info(
'Started %s on device [%s]', self.__class__.__name__, self.device
)
self._started_time = time.time()
while not self.should_stop and (
self.duration is None or time.time() - self._started_time < self.duration
):
self._wait_running()
if not converter:
self.wait_stop(0.1)
continue
if self.should_stop:
break
timeout = (
max(
0,
min(
self.duration - (time.time() - self._started_time),
self._converter_timeout,
),
)
if self.duration is not None
else self._converter_timeout
)
should_continue = self._process_converted_audio(
converter, timeout=timeout, out_f=out_f
)
if not should_continue:
break
def _process_converted_audio(
self, converter: AudioConverter, timeout: float, out_f: Optional[IO]
) -> bool:
"""
It reads the converted audio from the converter and passes it downstream.
:return: True if the process should continue, False if it should terminate.
"""
data = converter.read(timeout=timeout)
if not data:
return self._on_converter_timeout(converter)
self._on_audio_converted(data, out_f)
return True
def _on_converter_timeout(self, converter: AudioConverter) -> bool:
"""
Callback logic invoked if the converter times out.
:return: ``True`` (default) if the thread is supposed to continue,
``False`` if it should terminate.
"""
self.logger.debug('Timeout on converter %s', converter.__class__.__name__)
# Continue only if the converter hasn't terminated
return self._converter_terminated.is_set()
@override
def run(self):
"""
Wrapper for the main loop that initializes the converter and the stream.
"""
super().run()
self.paused_changed.clear()
try:
with self.open_converter() as converter, self._stream_type(
samplerate=self.sample_rate,
device=self.device,
channels=self.channels,
dtype=self.dtype,
latency=self.latency,
blocksize=self.blocksize,
**self._stream_args,
) as self.audio_stream, open(
self.outfile, 'wb'
) as out_f, self._audio_generator():
self.main(converter=converter, out_f=out_f)
except queue.Empty:
self.logger.warning(
'Audio callback timeout for %s', self.__class__.__name__
)
finally:
self.notify_stop()
@contextmanager
def _audio_generator(self) -> Generator[Optional[Thread], None, None]:
"""
:yield: A <Thread, Queue> pair where the thread generates raw audio
frames (as numpy arrays) that are sent to the specified queue.
"""
yield None
@contextmanager
def open_converter(self) -> Generator[Optional[AudioConverter], None, None]:
"""
Context manager for the converter process.
"""
if self._audio_converter_type is None:
yield None
return
assert not self._converter, 'A converter process is already running'
self._converter = self._audio_converter_type(
ffmpeg_bin=self.ffmpeg_bin,
sample_rate=self.sample_rate,
channels=self.channels,
volume=self.volume,
dtype=self.dtype,
chunk_size=self.blocksize,
on_exit=self._converter_terminated.set,
**self._converter_args,
)
self._converter.start()
yield self._converter
self._converter.stop()
self._converter.join(timeout=2)
self._converter = None
@contextmanager
def _change_state(self, state: AudioState, event_type: Type[SoundEvent]):
"""
Changes the state and it emits the specified event if the state has
actually changed.
It uses a context manager pattern, and everything in between will be
executed before the events are dispatched.
"""
with self._state_lock:
prev_state = self.state
self.state = state
yield
if prev_state != state:
self._notify(event_type)
def _notify(self, event_type: Type[SoundEvent], **kwargs):
"""
Notifies the specified event.
"""
get_bus().post(event_type(device=self.device, **kwargs))
def notify_start(self):
"""
Notifies the start event.
"""
with self._change_state(AudioState.RUNNING, self._started_event_type):
pass
def notify_stop(self):
"""
Notifies the stop event.
"""
with self._change_state(AudioState.STOPPED, self._stopped_event_type):
if self._converter:
self._converter.stop()
self.paused_changed.set()
self.paused_changed.clear()
def notify_pause(self):
"""
Notifies a pause toggle event.
"""
states = {
AudioState.PAUSED: AudioState.RUNNING,
AudioState.RUNNING: AudioState.PAUSED,
}
with self._state_lock:
new_state = states.get(self.state)
if not new_state:
return
event_type = (
self._paused_event_type
if new_state == AudioState.PAUSED
else self._resumed_event_type
)
with self._change_state(new_state, event_type):
self.paused_changed.set()
self.paused_changed.clear()
@property
def state(self):
"""
Thread-safe wrapper for the stream state.
"""
with self._state_lock:
return self._state
@state.setter
def state(self, value: AudioState):
"""
Thread-safe setter for the stream state.
"""
with self._state_lock:
self._state = value
def asdict(self) -> dict:
"""
Serialize the thread information.
"""
return {
'device': self.device,
'outfile': self.outfile,
'infile': self.infile,
'direction': self.direction,
'ffmpeg_bin': self.ffmpeg_bin,
'channels': self.channels,
'sample_rate': self.sample_rate,
'dtype': self.dtype,
'streaming': self.stream,
'duration': self.duration,
'blocksize': self.blocksize,
'latency': self.latency,
'redis_queue': self.redis_queue,
'audio_pass_through': self.audio_pass_through,
'state': self._state.value,
'volume': self.volume,
'started_time': datetime.fromtimestamp(self._started_time)
if self._started_time
else None,
'stream_index': self.stream_index,
'stream_name': self.stream_name,
}
# vim:sw=4:ts=4:et:

3
platypush/plugins/sound/_streams/_player/__init__.py Normal file
View file

@ -0,0 +1,3 @@
from ._base import AudioPlayer
__all__ = ['AudioPlayer']

110
platypush/plugins/sound/_streams/_player/_base.py Normal file
View file

@ -0,0 +1,110 @@
from abc import ABC
from typing import IO, Iterable, List, Optional, Self, Type, Union
from typing_extensions import override
import numpy as np
import sounddevice as sd
from platypush.message.event.sound import (
SoundPlaybackPausedEvent,
SoundPlaybackResumedEvent,
SoundPlaybackStartedEvent,
SoundPlaybackStoppedEvent,
)
from ..._converters import RawOutputAudioConverter
from ..._model import StreamType
from .._base import AudioThread
class AudioPlayer(AudioThread, ABC):
"""
Base ``AudioPlayer`` class.
An ``AudioPlayer`` thread is responsible for playing audio (either from a
file/URL or from a synthetic source) to an output device, writing it to the
converter process and dispatching the converted audio to the registered
consumers.
"""
def __init__(
self, *args, sound: Optional[Union[dict, Iterable[dict]]] = None, **kwargs
):
super().__init__(*args, **kwargs)
self.sound = sound
@classmethod
def build(
cls,
infile: Optional[str] = None,
sound: Optional[Union[dict, Iterable[dict]]] = None,
**kwargs,
) -> Self:
from ._resource import AudioResourcePlayer
from ._synth import AudioSynthPlayer, Sound
if infile:
return AudioResourcePlayer(infile=infile, **kwargs)
if sound:
sounds: List[dict] = ( # type: ignore
[sound] if isinstance(sound, dict) else sound
)
return AudioSynthPlayer(sounds=[Sound.build(**s) for s in sounds], **kwargs)
raise AssertionError('Either infile or url must be specified')
@property
@override
def direction(self) -> StreamType:
return StreamType.OUTPUT
@override
def _on_converter_timeout(self, *_, **__) -> bool:
return False # break
@property
@override
def _stream_type(self) -> Type[sd.RawOutputStream]:
return sd.RawOutputStream
@property
@override
def _audio_converter_type(self) -> Type[RawOutputAudioConverter]:
return RawOutputAudioConverter
@override
def _on_audio_converted(self, data: bytes, out_f: Optional[IO] = None):
if self.audio_stream:
self.audio_stream.write(
np.asarray(
self.gain
* np.frombuffer(data, dtype=self.dtype).reshape(-1, self.channels),
dtype=self.dtype,
)
)
super()._on_audio_converted(data, out_f)
@property
@override
def _started_event_type(self) -> Type[SoundPlaybackStartedEvent]:
return SoundPlaybackStartedEvent
@property
@override
def _stopped_event_type(self) -> Type[SoundPlaybackStoppedEvent]:
return SoundPlaybackStoppedEvent
@property
@override
def _paused_event_type(self) -> Type[SoundPlaybackPausedEvent]:
return SoundPlaybackPausedEvent
@property
@override
def _resumed_event_type(self) -> Type[SoundPlaybackResumedEvent]:
return SoundPlaybackResumedEvent
# vim:sw=4:ts=4:et:

39
platypush/plugins/sound/_streams/_player/_resource.py Normal file
View file

@ -0,0 +1,39 @@
from typing import Optional, Type
from typing_extensions import override
from platypush.message.event.sound import SoundEvent
from ..._converters import RawOutputAudioFromFileConverter
from ._base import AudioPlayer
class AudioResourcePlayer(AudioPlayer):
"""
A ``AudioResourcePlayer`` thread is responsible for playing an audio
resource - either a file or a URL.
"""
@property
@override
def _audio_converter_type(self) -> Type[RawOutputAudioFromFileConverter]:
return RawOutputAudioFromFileConverter
@property
@override
def _converter_args(self) -> dict:
return {
'infile': self.infile,
**super()._converter_args,
}
@property
@override
def _converter_stdin(self) -> Optional[int]:
return None
@override
def _notify(self, event_type: Type[SoundEvent], **kwargs):
return super()._notify(event_type, resource=self.infile, **kwargs)
# vim:sw=4:ts=4:et:

4
platypush/plugins/sound/_streams/_player/_synth/__init__.py Normal file
View file

@ -0,0 +1,4 @@
from ._player import AudioSynthPlayer
from ._sound import Sound
__all__ = ['AudioSynthPlayer', 'Sound']

79
platypush/plugins/sound/_streams/_player/_synth/_base.py Normal file
View file

@ -0,0 +1,79 @@
from abc import ABC, abstractmethod
from typing import Optional, Tuple
import numpy as np
from numpy.typing import NDArray
from ._parser import SoundParser
class SoundBase(SoundParser, ABC):
"""
Base class for synthetic sounds and mixes.
"""
def __init__(self, *args, volume: float = 100, **kwargs) -> None:
super().__init__(*args, **kwargs)
self.volume = volume
@property
def gain(self) -> float:
return self.volume / 100
@gain.setter
def gain(self, value: float):
self.volume = value * 100
@abstractmethod
def get_wave(
self,
sample_rate: float,
t_start: float = 0,
t_end: float = 0,
**_,
) -> NDArray[np.floating]:
"""
Get the wave binary data associated to this sound
:param t_start: Start offset for the wave in seconds. Default: 0
:param t_end: End offset for the wave in seconds. Default: 0
:param sample_rate: Audio sample rate. Default: 44100 Hz
:returns: A ``numpy.ndarray[(t_end-t_start)*sample_rate, 1]``
with the raw float values
"""
raise NotImplementedError()
def fft(
self,
sample_rate: float,
t_start: float = 0.0,
t_end: float = 0.0,
freq_range: Optional[Tuple[float, float]] = None,
freq_buckets: Optional[int] = None,
) -> NDArray[np.floating]:
"""
Get the real part of the Fourier transform associated to a time-bounded
sample of this sound.
:param t_start: Start offset for the wave in seconds. Default: 0
:param t_end: End offset for the wave in seconds. Default: 0
:param sample_rate: Audio sample rate. Default: 44100 Hz
:param freq_range: FFT frequency range. Default: ``(0, sample_rate/2)``
(see`Nyquist-Shannon sampling theorem
<https://en.wikipedia.org/wiki/Nyquist%E2%80%93Shannon_sampling_theorem>`_)
:param freq_buckets: Number of buckets to subdivide the frequency range.
Default: None
:returns: A numpy.ndarray[freq_range,1] with the raw float values
"""
if not freq_range:
freq_range = (0, int(sample_rate / 2))
wave = self.get_wave(t_start=t_start, t_end=t_end, sample_rate=sample_rate)
fft = np.fft.fft(wave.reshape(len(wave)))
fft = fft.real[freq_range[0] : freq_range[1]]
if freq_buckets is not None:
fft = np.histogram(fft, bins=freq_buckets)[0]
return fft

101
platypush/plugins/sound/_streams/_player/_synth/_generator.py Normal file
View file

@ -0,0 +1,101 @@
from logging import getLogger
from queue import Full, Queue
from threading import Thread
from time import time
from typing import Any, Callable, Optional
import numpy as np
from numpy.typing import NDArray
from ._mix import Mix
class AudioGenerator(Thread):
"""
The ``AudioGenerator`` class is a thread that generates synthetic raw audio
waves and dispatches them to a queue that can be consumed by other players,
streamers and converters.
"""
def __init__(
self,
*args,
audio_queue: Queue[NDArray[np.number]],
mix: Mix,
blocksize: int,
sample_rate: int,
queue_timeout: Optional[float] = None,
should_stop: Callable[[], bool] = lambda: False,
wait_running: Callable[[], Any] = lambda: None,
on_stop: Callable[[], Any] = lambda: None,
**kwargs,
):
super().__init__(*args, **kwargs)
self._audio_queue = audio_queue
self._t_start: float = 0
self._blocksize: int = blocksize
self._sample_rate: int = sample_rate
self._blocktime = self._blocksize / self._sample_rate
self._should_stop = should_stop
self._queue_timeout = queue_timeout
self._wait_running = wait_running
self._on_stop = on_stop
self.mix = mix
self.logger = getLogger(__name__)
def _next_t(self, t: float) -> float:
"""
Calculates the next starting time for the wave function.
"""
return (
min(t + self._blocktime, self._duration)
if self._duration is not None
else t + self._blocktime
)
def should_stop(self) -> bool:
"""
Stops if the upstream dependencies have signalled to stop or if the
duration is set and we have reached it.
"""
return self._should_stop() or (
self._duration is not None and time() - self._t_start >= self._duration
)
@property
def _duration(self) -> Optional[float]:
"""
Proxy to the mix object's duration.
"""
return self.mix.duration()
def run(self):
super().run()
self._t_start = time()
t = 0
while not self.should_stop():
self._wait_running()
if self.should_stop():
break
next_t = self._next_t(t)
try:
data = self.mix.get_wave(
t_start=t, t_end=next_t, sample_rate=self._sample_rate
)
except Exception as e:
self.logger.warning('Could not generate the audio wave: %s', e)
break
try:
self._audio_queue.put(data, timeout=self._queue_timeout)
t = next_t
except Full:
self.logger.warning(
'The processing queue is full: either the audio consumer is stuck, '
'or you may want to increase queue_size'
)
self._on_stop()

115
platypush/plugins/sound/_streams/_player/_synth/_mix.py Normal file
View file

@ -0,0 +1,115 @@
import json
import logging
from typing import List, Tuple, Union
from typing_extensions import override
import numpy as np
from numpy.typing import DTypeLike, NDArray
from ...._utils import convert_nd_array
from ._base import SoundBase
from ._sound import Sound
class Mix(SoundBase):
"""
This class models a set of mixed :class:`._sound.Sound` instances that can be played
through an audio stream to an audio device
"""
def __init__(self, *sounds, channels: int, dtype: DTypeLike, **kwargs):
super().__init__(**kwargs)
self._sounds: List[Sound] = []
self.logger = logging.getLogger(__name__)
self.channels = channels
self.dtype = np.dtype(dtype)
for sound in sounds:
self.add(sound)
def __iter__(self):
"""
Iterate over the object's attributes and return key-pair values.
"""
for sound in self._sounds:
yield dict(sound)
def __str__(self):
"""
Return a JSON string representation of the object.
"""
return json.dumps(list(self))
def add(self, *sounds: Union[Sound, dict]):
"""
Add one or more sounds to the mix.
"""
self._sounds += [Sound.build(sound) for sound in sounds]
def remove(self, *sound_indices: int):
"""
Remove one or more sounds from the mix.
"""
assert self._sounds and all(
0 <= sound_index < len(sound_indices) for sound_index in sound_indices
), f'Sound indices must be between 0 and {len(self._sounds) - 1}'
for sound_index in sound_indices[::-1]:
self._sounds.pop(sound_index)
@override
def get_wave(
self,
sample_rate: float,
t_start: float = 0,
t_end: float = 0,
normalize_range: Tuple[float, float] = (-1.0, 1.0),
on_clip: str = 'scale',
**_,
) -> NDArray[np.number]:
wave = None
for sound in self._sounds:
sound_wave = sound.get_wave(
t_start=t_start, t_end=t_end, sample_rate=sample_rate
)
if wave is None:
wave = sound_wave
else:
wave += sound_wave
if wave is not None and len(wave):
scale_factor = (normalize_range[1] - normalize_range[0]) / (
wave.max() - wave.min()
)
if scale_factor < 1.0: # Wave clipping
if on_clip == 'scale':
wave = scale_factor * wave
elif on_clip == 'clip':
wave[wave < normalize_range[0]] = normalize_range[0]
wave[wave > normalize_range[1]] = normalize_range[1]
else:
raise RuntimeError(
'Supported values for "on_clip": ' + '"scale" or "clip"'
)
assert wave is not None
return convert_nd_array(self.gain * wave, dtype=self.dtype)
def duration(self):
"""
:returns: The duration of the mix in seconds as duration of its longest
sample, or None if the mixed sample have no duration set
"""
# If any sound has no duration specified, then the resulting mix will
# have no duration as well.
if any(sound.duration is None for sound in self._sounds):
return None
return max(((sound.duration or 0) + sound.delay for sound in self._sounds))
# vim:sw=4:ts=4:et:

79
platypush/plugins/sound/_streams/_player/_synth/_output.py Normal file
View file

@ -0,0 +1,79 @@
from logging import getLogger
from queue import Empty, Queue
from typing import Callable, Optional
import sounddevice as sd
import numpy as np
from numpy.typing import NDArray
# pylint: disable=too-few-public-methods
class AudioOutputCallback:
"""
The ``AudioSynthOutput`` is a functor that wraps the ``sounddevice.Stream``
callback and writes raw audio data to the audio device.
"""
def __init__(
self,
*args,
audio_queue: Queue[NDArray[np.number]],
channels: int,
blocksize: int,
should_stop: Callable[[], bool] = lambda: False,
is_paused: Callable[[], bool] = lambda: False,
queue_timeout: Optional[float] = None,
**kwargs,
):
super().__init__(*args, **kwargs)
self._audio_queue = audio_queue
self._channels = channels
self._blocksize = blocksize
self._should_stop = should_stop
self._is_paused = is_paused
self._queue_timeout = queue_timeout
self.logger = getLogger(__name__)
def _check_status(self, frames: int, status):
"""
Checks the current status of the audio callback and raises errors if
the processing shouldn't continue.
"""
if self._should_stop():
raise sd.CallbackStop
assert frames == self._blocksize, (
f'Received {frames} frames, expected blocksize is {self._blocksize}',
)
assert not status.output_underflow, 'Output underflow: increase blocksize?'
assert not status, f'Audio callback failed: {status}'
def _audio_callback(self, outdata: NDArray[np.number], frames: int, status):
if self._is_paused():
return
self._check_status(frames, status)
try:
data = self._audio_queue.get_nowait()
except Empty as e:
raise (
sd.CallbackStop
if self._should_stop()
else AssertionError('Buffer is empty: increase buffersize?')
) from e
if data.shape[0] == 0:
raise sd.CallbackStop
audio_length = min(len(data), len(outdata))
outdata[:audio_length] = data[:audio_length]
# _ = time
def __call__(self, outdata: NDArray[np.number], frames: int, _, status):
try:
self._audio_callback(outdata, frames, status)
except AssertionError as e:
self.logger.warning(str(e))

111
platypush/plugins/sound/_streams/_player/_synth/_parser.py Normal file
View file

@ -0,0 +1,111 @@
import math
import re
from typing import Optional, Union
class SoundParser:
"""
A utility mixin with some methods to parse and convert sound information -
e.g. MIDI notes from strings, MIDI notes to frequencies, and the other way
around.
"""
_DEFAULT_A4_FREQUENCY = 440.0
_MIDI_NOTE_REGEX = re.compile(r'^([A-G])([#b]?)(-?[0-9]+)$')
_MID_A_MIDI_NOTE = 69
_NOTE_OFFSETS = {
'C': 0,
'C#': 1,
'Db': 1,
'D': 2,
'D#': 3,
'Eb': 3,
'E': 4,
'F': 5,
'F#': 6,
'Gb': 6,
'G': 7,
'G#': 8,
'Ab': 8,
'A': 9,
'A#': 10,
'Bb': 10,
'B': 11,
}
_ALTERATION_OFFSETS = {
'b': -1,
'': 0,
'#': 1,
}
def __init__(self, *_, ref_frequency: float = _DEFAULT_A4_FREQUENCY, **__) -> None:
self._ref_frequency = ref_frequency
@staticmethod
def _get_alteration_offset(alt: str) -> int:
"""
Calculate the MIDI note offset given by its reported sharp/flat alteration.
"""
if alt == '#':
return 1
if alt == 'b':
return -1
return 0
@classmethod
def get_midi_note(cls, note: Union[str, int]) -> int:
"""
Convert a MIDI note given as input (either an integer or a string like
'C4') to a MIDI note number.
:raise: ValueError
"""
if isinstance(note, str):
note = note[:1].upper() + note[1:]
m = cls._MIDI_NOTE_REGEX.match(note)
if not m:
raise ValueError(f'Invalid MIDI note: {note}')
base_note, alteration, octave = m.groups()
octave = int(octave)
note_offset = cls._NOTE_OFFSETS[base_note] + cls._get_alteration_offset(
alteration
)
octave_offset = (octave + 1) * 12
note = octave_offset + note_offset
if isinstance(note, int):
if not 0 <= note <= 127:
raise ValueError(f'MIDI note out of range: {note}')
return note
raise ValueError(f'Invalid MIDI note: {note}')
def note_to_freq(
self, midi_note: Union[int, str], ref_frequency: Optional[float] = None
):
"""
Converts a MIDI note to its frequency in Hz
:param midi_note: MIDI note to convert
:param ref_frequency: Reference A4 frequency override (default: 440 Hz).
"""
note = self.get_midi_note(midi_note)
return (2.0 ** ((note - self._MID_A_MIDI_NOTE) / 12.0)) * (
ref_frequency or self._ref_frequency
)
def freq_to_note(self, frequency: float, ref_frequency: Optional[float] = None):
"""
Converts a frequency in Hz to its closest MIDI note
:param frequency: Frequency in Hz
:param ref_frequency: Reference A4 frequency override (default: 440 Hz).
"""
std_freq = ref_frequency or self._ref_frequency
return int(12.0 * math.log(frequency / std_freq, 2) + self._MID_A_MIDI_NOTE)

125
platypush/plugins/sound/_streams/_player/_synth/_player.py Normal file
View file

@ -0,0 +1,125 @@
from contextlib import contextmanager
from queue import Queue
from threading import Event
from typing import Any, Generator, Iterable, Optional, Type
from typing_extensions import override
import numpy as np
import sounddevice as sd
from numpy.typing import DTypeLike, NDArray
from ...._model import AudioState
from ..._player import AudioPlayer
from ._generator import AudioGenerator
from ._mix import Mix
from ._output import AudioOutputCallback
from ._sound import Sound
class AudioSynthPlayer(AudioPlayer):
"""
The ``AudioSynthPlayer`` can play synthetic sounds (specified either by MIDI
note or raw frequency) to an audio device.
"""
def __init__(
self,
*args,
volume: float,
channels: int,
dtype: DTypeLike,
sounds: Optional[Iterable[Sound]] = None,
**kwargs
):
sounds = sounds or []
self.mix = Mix(*sounds, volume=volume, channels=channels, dtype=dtype)
super().__init__(*args, volume=volume, channels=channels, dtype=dtype, **kwargs)
self._generator_stopped = Event()
self._completed_callback_event = Event()
self._audio_queue: Queue[NDArray[np.number]] = Queue(
maxsize=self.queue_size or 0
)
@property
@override
def _stream_type(self) -> Type[sd.OutputStream]:
return sd.OutputStream
@property
@override
def _audio_converter_type(self) -> None:
pass
def __setattr__(self, __name: str, __value: Any):
"""
Make sure that the relevant attributes are synchronized to the mix
object upon set/update.
"""
if __name == 'volume':
# Propagate the volume changes to the mix object.
self.mix.volume = __value
return super().__setattr__(__name, __value)
@override
def _on_converter_timeout(self, *_, **__) -> bool:
"""
Don't break the audio stream if the output converter failed
"""
return True
@property
@override
def _stream_args(self) -> dict:
"""
Register an :class:`.AudioOutputCallback` to fill up the audio buffers.
"""
return {
'callback': AudioOutputCallback(
audio_queue=self._audio_queue,
channels=self.channels,
blocksize=self.blocksize,
queue_timeout=self._queue_timeout,
should_stop=lambda: self.should_stop
or self._generator_stopped.is_set(),
is_paused=lambda: self.state == AudioState.PAUSED,
),
'finished_callback': self._completed_callback_event.set,
**super()._stream_args,
}
@property
def _queue_timeout(self) -> float:
"""
Estimated max read/write timeout on the audio queue.
"""
return self.blocksize * (self.queue_size or 5) / self.sample_rate
@override
@contextmanager
def _audio_generator(self) -> Generator[AudioGenerator, None, None]:
stop_generator = Event()
gen = AudioGenerator(
audio_queue=self._audio_queue,
mix=self.mix,
blocksize=self.blocksize,
sample_rate=self.sample_rate,
queue_timeout=self._queue_timeout,
should_stop=lambda: self.should_stop or stop_generator.is_set(),
wait_running=self._wait_running,
on_stop=self._on_stop,
)
self._generator_stopped.clear()
gen.start()
yield gen
stop_generator.set()
gen.join()
def _on_stop(self):
self._generator_stopped.set()
self.notify_stop()
# vim:sw=4:ts=4:et:

225
platypush/plugins/sound/_streams/_player/_synth/_sound.py Normal file
View file

@ -0,0 +1,225 @@
from enum import Enum
import json
from typing import Final, Optional, Tuple, Union
from typing_extensions import override
import numpy as np
from numpy.typing import NDArray
from ._base import SoundBase
class WaveShape(Enum):
"""
Supported audio wave shapes.
"""
SIN = 'sin'
SQUARE = 'square'
SAWTOOTH = 'sawtooth'
TRIANG = 'triang'
class Sound(SoundBase):
"""
Models a basic synthetic sound that can be played through an audio device
"""
_DEFAULT_MID_A_FREQUENCY: Final[float] = 440.0
def __init__(
self,
*args,
midi_note: Optional[Union[str, int]] = None,
frequency: Optional[float] = None,
phase: float = 0,
duration: Optional[float] = None,
delay: float = 0,
shape: WaveShape = WaveShape.SIN,
**kwargs,
):
"""
You can construct a sound either from a MIDI note or a base frequency,
as well as the shape of the output wave.
:param midi_note: MIDI note code, see `this chart
<https://newt.phys.unsw.edu.au/jw/graphics/notes.GIF>`_.
:param frequency: Sound base frequency in Hz
:param phase: Wave phase shift as a multiple of pi (default: 0.0)
:param duration: Note duration in seconds. Default: keep until
release/pause/stop
:param delay: Sound delay in seconds, calculated from the moment the
command execution starts. Default: 0.
:param shape: Wave shape. Possible values: "``sin``", "``square``",
"``sawtooth``" or "``triang``" (see :class:`WaveShape`).
Default: "``sin``"
"""
super().__init__(*args, **kwargs)
invalid_request = RuntimeError(
'Please specify either a MIDI note or a base frequency'
)
if midi_note and frequency:
raise invalid_request
if midi_note:
self.midi_note = self.get_midi_note(midi_note)
self.frequency = self.note_to_freq(midi_note=midi_note)
elif frequency:
self.frequency = frequency
self.midi_note = self.freq_to_note(frequency=frequency)
else:
raise invalid_request
self.phase = phase
self.duration = duration
self.delay = delay
self.shape = WaveShape(shape)
def _get_left_audio_pad(
self, sample_rate: float, t_start: float, t_end: float
) -> int:
"""
Get the size of the audio wave left zero-pad given in function of its
``delay``, ``sample_rate``, ``t_start`` and ``t_end``.
"""
return round(max(0, min(t_end, self.delay) - t_start) * sample_rate)
def _get_right_audio_pad(
self, sample_rate: float, t_start: float, t_end: float
) -> int:
"""
Get the size of the audio wave right zero-pad given its declared
``delay`` in function of ``t_start`` and ``t_end``.
"""
if not self.duration:
return 0
duration = self.delay + self.duration
if t_end <= duration:
return 0
return round((t_end - max(t_start, duration)) * sample_rate)
def _get_audio_pad(
self, sample_rate: float, t_start: float, t_end: float
) -> Tuple[NDArray[np.floating], NDArray[np.floating]]:
"""
Return the left and right audio pads for a given audio length as a
``(left, right)`` tuple of numpy zero-filled arrays.
"""
return tuple(
np.zeros([pad_size, 1])
for pad_size in (
self._get_left_audio_pad(
sample_rate=sample_rate, t_start=t_start, t_end=t_end
),
self._get_right_audio_pad(
sample_rate=sample_rate, t_start=t_start, t_end=t_end
),
)
)
def _generate_wave(self, x: NDArray[np.floating]):
"""
Generate a raw audio wave as a numpy array of floating between -1 and 1
given ``x`` as a set of timestamp samples.
"""
if self.shape in (WaveShape.SIN, WaveShape.SQUARE):
wave = np.sin((2 * np.pi * self.frequency * x) + np.pi * self.phase)
if self.shape == WaveShape.SQUARE:
wave[wave < 0] = -0.95
wave[wave >= 0] = 0.95
elif self.shape in (WaveShape.SAWTOOTH, WaveShape.TRIANG):
wave = 2 * (self.frequency * x - np.floor(0.5 + self.frequency * x))
if self.shape == WaveShape.TRIANG:
wave = 2 * np.abs(wave) - 1
else:
raise RuntimeError(
f'Unsupported wave shape: {self.shape}. '
f'Supported values: {[s.value for s in WaveShape]}'
)
return wave
@override
def get_wave(
self,
sample_rate: float,
t_start: float = 0,
t_end: float = 0,
**_,
) -> NDArray[np.floating]:
"""
Get the wave binary data associated to this sound
:param t_start: Start offset for the wave in seconds. Default: 0
:param t_end: End offset for the wave in seconds. Default: 0
:param sample_rate: Audio sample rate. Default: 44100 Hz
:returns: A ``numpy.ndarray[(t_end-t_start)*sample_rate, 1]``
with the raw float values
"""
assert self.frequency is not None, 'The sound has no associated base frequency'
if t_start > t_end:
return np.array([])
left_pad, right_pad = self._get_audio_pad(
sample_rate=sample_rate, t_start=t_start, t_end=t_end
)
t_start = min(t_end, t_start + (left_pad.shape[0] / sample_rate))
t_end = max(t_start, t_end - (right_pad.shape[0] / sample_rate))
actual_n_samples = abs(round((t_end - t_start) * sample_rate))
wave_length = max(t_start, self.delay - t_start)
if self.duration is not None:
wave_length = min(wave_length, self.duration - self.delay)
x = np.linspace(
max(t_start, self.delay - t_start),
t_end,
actual_n_samples,
).reshape(-1, 1)
return self.gain * np.array(
(
*left_pad,
*self._generate_wave(x),
*right_pad,
)
)
def __iter__(self):
"""
Iterates over the sound's attributes and returns key-value pairs.
"""
for attr in ['midi_note', 'frequency', 'volume', 'duration', 'ref_frequency']:
yield attr, getattr(self, attr)
def __str__(self):
"""
:return: A JSON-string representation of the sound dictionary.
"""
return json.dumps(dict(self))
@classmethod
def build(cls, *args, **kwargs) -> "Sound":
"""
Construct a sound object either from a JSON representation or a
key-value representation
"""
if args:
if isinstance(args[0], cls):
return args[0]
if isinstance(args[0], str):
kwargs = json.loads(args[0])
elif isinstance(args[0], dict):
kwargs = args[0]
if kwargs:
return Sound(**kwargs)
raise RuntimeError(f'Usage: {__doc__}')

106
platypush/plugins/sound/_streams/_recorder.py Normal file
View file

@ -0,0 +1,106 @@
from typing import Type
from typing_extensions import override
import sounddevice as sd
from platypush.message.event.sound import (
SoundRecordingPausedEvent,
SoundRecordingResumedEvent,
SoundRecordingStartedEvent,
SoundRecordingStoppedEvent,
)
from .._converters import RawInputAudioConverter
from .._model import AudioState, StreamType
from ._base import AudioThread
class AudioRecorder(AudioThread):
"""
The ``AudioRecorder`` thread is responsible for recording audio from the
input device, writing it to the converter process and dispatch the
converted audio to the registered consumers.
"""
def __init__(self, *args, output_format: str, **kwargs):
super().__init__(*args, **kwargs)
self.output_format = output_format
@property
@override
def direction(self) -> StreamType:
return StreamType.INPUT
@override
def _audio_callback(self):
# _ = frames
# __ = time
def callback(indata, outdata, _, __, status):
if self.state != AudioState.RUNNING:
return
if status:
self.logger.warning('Recording callback status: %s', status)
if not self._converter:
self.logger.warning(
'The ffmpeg converter process has already terminated'
)
self.notify_stop()
raise sd.CallbackStop
try:
self._converter.write(indata.tobytes())
except AssertionError as e:
self.logger.warning('Audio converter callback error: %s', e)
self.state = AudioState.STOPPED
return
if self.audio_pass_through:
outdata[:] = indata
return callback
@property
@override
def _audio_converter_type(self) -> Type[RawInputAudioConverter]:
return RawInputAudioConverter
@property
@override
def _started_event_type(self) -> Type[SoundRecordingStartedEvent]:
return SoundRecordingStartedEvent
@property
@override
def _stopped_event_type(self) -> Type[SoundRecordingStoppedEvent]:
return SoundRecordingStoppedEvent
@property
@override
def _paused_event_type(self) -> Type[SoundRecordingPausedEvent]:
return SoundRecordingPausedEvent
@property
@override
def _resumed_event_type(self) -> Type[SoundRecordingResumedEvent]:
return SoundRecordingResumedEvent
@property
@override
def _converter_args(self) -> dict:
return {
'format': self.output_format,
**super()._converter_args,
}
@property
@override
def _stream_args(self) -> dict:
return {
'callback': self._audio_callback(),
**super()._stream_args,
}
# vim:sw=4:ts=4:et:

3
platypush/plugins/sound/_utils/__init__.py Normal file
View file

@ -0,0 +1,3 @@
from ._convert import convert_nd_array
__all__ = ["convert_nd_array"]

28
platypush/plugins/sound/_utils/_convert.py Normal file
View file

@ -0,0 +1,28 @@
import numpy as np
from numpy.typing import DTypeLike, NDArray
def convert_nd_array( # pylint: disable=too-many-return-statements
wave: NDArray[np.floating], dtype: DTypeLike
) -> NDArray[np.number]:
"""
Given a wave as a series of floating point numbers, convert them to the
appropriate data type.
"""
t = np.dtype(dtype)
if t in {np.float16, np.float32, np.float64}:
return wave.astype(t)
if t == np.int8:
return (wave * 2**8).astype(t)
if t == np.uint8:
return ((wave + 1) * 2**8).astype(t)
if t == np.int16:
return (wave * 2**15).astype(t)
if t == np.uint16:
return ((wave + 1) * 2**16).astype(t)
if t == np.int32:
return (wave * 2**31).astype(t)
if t == np.uint32:
return ((wave + 1) * 2**32).astype(t)
raise AssertionError(f'Unsupported dtype: {dtype}')

405
platypush/plugins/sound/core.py
View file

@ -1,405 +0,0 @@
import enum
import logging
import json
import math
class WaveShape(enum.Enum):
SIN = 'sin'
SQUARE = 'square'
SAWTOOTH = 'sawtooth'
TRIANG = 'triang'
class Sound:
"""
Models a basic synthetic sound that can be played through an audio device
"""
STANDARD_A_FREQUENCY = 440.0
STANDARD_A_MIDI_NOTE = 69
_DEFAULT_BLOCKSIZE = 1024
_DEFAULT_SYNTH_BUFSIZE = 2
_DEFAULT_FILE_BUFSIZE = 20
_DEFAULT_SAMPLERATE = 44100
midi_note = None
frequency = None
phase = 0.0
gain = 1.0
duration = None
shape = None
def __init__(
self,
midi_note=midi_note,
frequency=None,
phase=phase,
gain=gain,
duration=duration,
shape=WaveShape.SIN,
A_frequency=STANDARD_A_FREQUENCY,
):
"""
You can construct a sound either from a MIDI note or a base frequency
:param midi_note: MIDI note code, see
https://newt.phys.unsw.edu.au/jw/graphics/notes.GIF
:type midi_note: int
:param frequency: Sound base frequency in Hz
:type frequency: float
:param phase: Wave phase shift as a multiple of pi (default: 0.0)
:type phase: float
:param gain: Note gain/volume between 0.0 and 1.0 (default: 1.0)
:type gain: float
:param duration: Note duration in seconds. Default: keep until
release/pause/stop
:type duration: float
:param shape: Wave shape. Possible values: "``sin``", "``square``",
"``sawtooth``" or "``triang``" (see :class:`WaveSound`).
Default: "``sin``"
:type shape: str
:param A_frequency: Reference A4 frequency (default: 440 Hz)
:type A_frequency: float
"""
if midi_note and frequency:
raise RuntimeError(
'Please specify either a MIDI note or a base ' + 'frequency'
)
if midi_note:
self.midi_note = midi_note
self.frequency = self.note_to_freq(
midi_note=midi_note, A_frequency=A_frequency
)
elif frequency:
self.frequency = frequency
self.midi_note = self.freq_to_note(
frequency=frequency, A_frequency=A_frequency
)
else:
raise RuntimeError(
'Please specify either a MIDI note or a base ' + 'frequency'
)
self.phase = phase
self.gain = gain
self.duration = duration
self.shape = WaveShape(shape)
@classmethod
def note_to_freq(cls, midi_note, A_frequency=STANDARD_A_FREQUENCY):
"""
Converts a MIDI note to its frequency in Hz
:param midi_note: MIDI note to convert
:type midi_note: int
:param A_frequency: Reference A4 frequency (default: 440 Hz)
:type A_frequency: float
"""
return (2.0 ** ((midi_note - cls.STANDARD_A_MIDI_NOTE) / 12.0)) * A_frequency
@classmethod
def freq_to_note(cls, frequency, A_frequency=STANDARD_A_FREQUENCY):
"""
Converts a frequency in Hz to its closest MIDI note
:param frequency: Frequency in Hz
:type frequency: float
:param A_frequency: Reference A4 frequency (default: 440 Hz)
:type A_frequency: float
"""
# TODO return also the offset in % between the provided frequency
# and the standard MIDI note frequency
return int(
12.0 * math.log(frequency / A_frequency, 2) + cls.STANDARD_A_MIDI_NOTE
)
def get_wave(self, t_start=0.0, t_end=0.0, samplerate=_DEFAULT_SAMPLERATE):
"""
Get the wave binary data associated to this sound
:param t_start: Start offset for the wave in seconds. Default: 0
:type t_start: float
:param t_end: End offset for the wave in seconds. Default: 0
:type t_end: float
:param samplerate: Audio sample rate. Default: 44100 Hz
:type samplerate: int
:returns: A ``numpy.ndarray[(t_end-t_start)*samplerate, 1]``
with the raw float values
"""
import numpy as np
x = np.linspace(t_start, t_end, int((t_end - t_start) * samplerate))
x = x.reshape(len(x), 1)
if self.shape == WaveShape.SIN or self.shape == WaveShape.SQUARE:
wave = np.sin((2 * np.pi * self.frequency * x) + np.pi * self.phase)
if self.shape == WaveShape.SQUARE:
wave[wave < 0] = -1
wave[wave >= 0] = 1
elif self.shape == WaveShape.SAWTOOTH or self.shape == WaveShape.TRIANG:
wave = 2 * (self.frequency * x - np.floor(0.5 + self.frequency * x))
if self.shape == WaveShape.TRIANG:
wave = 2 * np.abs(wave) - 1
else:
raise RuntimeError('Unsupported wave shape: {}'.format(self.shape))
return self.gain * wave
def fft(
self,
t_start=0.0,
t_end=0.0,
samplerate=_DEFAULT_SAMPLERATE,
freq_range=None,
freq_buckets=None,
):
"""
Get the real part of the Fourier transform associated to a time-bounded
sample of this sound
:param t_start: Start offset for the wave in seconds. Default: 0
:type t_start: float
:param t_end: End offset for the wave in seconds. Default: 0
:type t_end: float
:param samplerate: Audio sample rate. Default: 44100 Hz
:type samplerate: int
:param freq_range: FFT frequency range. Default: ``(0, samplerate/2)``
(see`Nyquist-Shannon sampling theorem
<https://en.wikipedia.org/wiki/Nyquist%E2%80%93Shannon_sampling_theorem>`_)
:type freq_range: list or tuple with 2 int elements (range)
:param freq_buckets: Number of buckets to subdivide the frequency range.
Default: None
:type freq_buckets: int
:returns: A numpy.ndarray[freq_range,1] with the raw float values
"""
import numpy as np
if not freq_range:
freq_range = (0, int(samplerate / 2))
wave = self.get_wave(t_start=t_start, t_end=t_end, samplerate=samplerate)
fft = np.fft.fft(wave.reshape(len(wave)))
fft = fft.real[freq_range[0] : freq_range[1]]
if freq_buckets is not None:
fft = np.histogram(fft, bins=freq_buckets)
return fft
def __iter__(self):
for attr in ['midi_note', 'frequency', 'gain', 'duration']:
yield attr, getattr(self, attr)
def __str__(self):
return json.dumps(dict(self))
@classmethod
def build(cls, *args, **kwargs):
"""
Construct a sound object either from a JSON representation or a
key-value representation
"""
if args:
if isinstance(args[0], cls):
return args[0]
if isinstance(args[0], str):
kwargs = json.loads(args[0])
elif isinstance(args[0], dict):
kwargs = args[0]
if kwargs:
return Sound(**kwargs)
raise RuntimeError('Usage: {}'.format(__doc__))
class Mix:
"""
This class models a set of mixed :class:`Sound` instances that can be played
through an audio stream to an audio device
"""
_sounds = None
def __init__(self, *sounds):
self._sounds = []
self.logger = logging.getLogger(__name__)
for sound in sounds:
self.add(sound)
def __iter__(self):
for sound in self._sounds:
yield dict(sound)
def __str__(self):
return json.dumps(list(self))
def add(self, sound):
self._sounds.append(Sound.build(sound))
def remove(self, sound_index):
if sound_index >= len(self._sounds):
self.logger.error(
'No such sound index: {} in mix {}'.format(sound_index, list(self))
)
return
self._sounds.pop(sound_index)
# noinspection PyProtectedMember
def get_wave(
self,
t_start=0.0,
t_end=0.0,
normalize_range=(-1.0, 1.0),
on_clip='scale',
samplerate=Sound._DEFAULT_SAMPLERATE,
):
"""
Get the wave binary data associated to this mix
:param t_start: Start offset for the wave in seconds. Default: 0
:type t_start: float
:param t_end: End offset for the wave in seconds. Default: 0
:type t_end: float
:param normalize_range: Normalization range. If set the gain values of the
wave will be normalized to fit into the specified range if it
"clips" above or below. Default: ``(-1.0, 1.0)``
:type normalize_range: list[float]
:param on_clip: Action to take on wave clipping if ``normalize_range``
is set. Possible values: "``scale``" (scale down the frame to remove
the clipping) or "``clip``" (saturate the values above/below range).
Default: "``scale``".
:type on_clip: str
:param samplerate: Audio sample rate. Default: 44100 Hz
:type samplerate: int
:returns: A numpy.ndarray[n,1] with the raw float values
"""
wave = None
for sound in self._sounds:
sound_wave = sound.get_wave(
t_start=t_start, t_end=t_end, samplerate=samplerate
)
if wave is None:
wave = sound_wave
else:
wave += sound_wave
if normalize_range and len(wave):
scale_factor = (normalize_range[1] - normalize_range[0]) / (
wave.max() - wave.min()
)
if scale_factor < 1.0: # Wave clipping
if on_clip == 'scale':
wave = scale_factor * wave
elif on_clip == 'clip':
wave[wave < normalize_range[0]] = normalize_range[0]
wave[wave > normalize_range[1]] = normalize_range[1]
else:
raise RuntimeError(
'Supported values for "on_clip": ' + '"scale" or "clip"'
)
return wave
# noinspection PyProtectedMember
def fft(
self,
t_start=0.0,
t_end=0.0,
samplerate=Sound._DEFAULT_SAMPLERATE,
freq_range=None,
freq_buckets=None,
):
"""
Get the real part of the Fourier transform associated to a time-bounded
sample of this mix
:param t_start: Start offset for the wave in seconds. Default: 0
:type t_start: float
:param t_end: End offset for the wave in seconds. Default: 0
:type t_end: float
:param samplerate: Audio sample rate. Default: 44100 Hz
:type samplerate: int
:param freq_range: FFT frequency range. Default: ``(0, samplerate/2)``
(see `Nyquist-Shannon sampling theorem
<https://en.wikipedia.org/wiki/Nyquist%E2%80%93Shannon_sampling_theorem>`_)
:type freq_range: list or tuple with 2 int elements (range)
:param freq_buckets: Number of buckets to subdivide the frequency range.
Default: None
:type freq_buckets: int
:returns: A numpy.ndarray[freq_range,1] with the raw float values
"""
import numpy as np
if not freq_range:
freq_range = (0, int(samplerate / 2))
wave = self.get_wave(t_start=t_start, t_end=t_end, samplerate=samplerate)
fft = np.fft.fft(wave.reshape(len(wave)))
fft = fft.real[freq_range[0] : freq_range[1]]
if freq_buckets is not None:
fft = np.histogram(fft, bins=freq_buckets)
return fft
def duration(self):
"""
:returns: The duration of the mix in seconds as duration of its longest
sample, or None if the mixed sample have no duration set
"""
duration = 0
for sound in self._sounds:
if sound.duration is None:
return None
duration = max(duration, sound.duration)
return duration
# vim:sw=4:ts=4:et:

3
platypush/plugins/sound/manifest.yaml
View file

@ -1,15 +1,16 @@
manifest:
events:
platypush.message.event.sound.SoundPlaybackPausedEvent: on playback pause
platypush.message.event.sound.SoundPlaybackResumedEvent: on playback resume
platypush.message.event.sound.SoundPlaybackStartedEvent: on playback start
platypush.message.event.sound.SoundPlaybackStoppedEvent: on playback stop
platypush.message.event.sound.SoundRecordingPausedEvent: on recording pause
platypush.message.event.sound.SoundRecordingResumedEvent: on recording resumed
platypush.message.event.sound.SoundRecordingStartedEvent: on recording start
platypush.message.event.sound.SoundRecordingStoppedEvent: on recording stop
install:
pip:
- sounddevice
- soundfile
- numpy
apt:
- ffmpeg

2
setup.py
View file

@ -154,7 +154,7 @@ setup(
# Support for Chromecast plugin
'chromecast': ['pychromecast'],
# Support for sound devices
'sound': ['sounddevice', 'soundfile', 'numpy'],
'sound': ['sounddevice', 'numpy'],
# Support for web media subtitles
'subtitles': [
'webvtt-py',