platypush/platypush/plugins/tts/google.py

100 lines
3.2 KiB
Python

import os
import subprocess
import tempfile
from google.cloud import texttospeech
from platypush.plugins import Plugin, action
class TtsGooglePlugin(Plugin):
"""
Advanced text-to-speech engine that leverages the Google Cloud TTS API.
See https://cloud.google.com/text-to-speech/docs/quickstart-client-libraries#client-libraries-install-python
for how to enable the API on your account and get your credentials.
Requires:
* **google-cloud-texttospeech** - ``pip install google-cloud-texttospeech``
* **mplayer** - see your distribution docs on how to install the mplayer package
"""
def __init__(self, language='en-US', voice='en-US-Wavenet-C',
gender='FEMALE', credentials_file='~/.credentials/platypush/google/platypush-tts.json'):
"""
:param language: Language code, see https://cloud.google.com/text-to-speech/docs/basics for supported languages
:type language: str
:param voice: Voice type, see https://cloud.google.com/text-to-speech/docs/basics for supported voices
:type voice: str
:param gender: Voice gender (MALE, FEMALE or NEUTRAL)
:type gender: str
:param credentials_file: Where your GCloud credentials for TTS are stored, see https://cloud.google.com/text-to-speech/docs/basics
:type credentials_file: str
"""
super().__init__()
self.language = language
self.voice = voice
self.gender = getattr(texttospeech.enums.SsmlVoiceGender, gender.upper())
os.environ['GOOGLE_APPLICATION_CREDENTIALS'] = os.path.expanduser(credentials_file)
@action
def say(self, text, language=None, voice=None, gender=None):
"""
Say a phrase
:param text: Text to say
:type text: str
:param language: Language code override
:type language: str
:param voice: Voice type override
:type voice: str
:param gender: Gender override
:type gender: str
"""
client = texttospeech.TextToSpeechClient()
synthesis_input = texttospeech.types.SynthesisInput(text=text)
if language is None:
language = self.language
if gender is None:
gender = self.gender
else:
gender = getattr(texttospeech.enums.SsmlVoiceGender, gender.upper())
if voice is None:
voice = self.voice
voice = texttospeech.types.VoiceSelectionParams(
language_code=language, ssml_gender=gender,
name=voice)
audio_config = texttospeech.types.AudioConfig(
audio_encoding=texttospeech.enums.AudioEncoding.MP3)
response = client.synthesize_speech(synthesis_input, voice, audio_config)
with tempfile.NamedTemporaryFile() as f:
f.write(response.audio_content)
output = None
errors = []
cmd = ['mplayer -ao alsa -really-quiet -noconsolecontrols {}'
.format(f.name)]
try:
return subprocess.check_output(
cmd, stderr=subprocess.STDOUT, shell=True).decode('utf-8')
except subprocess.CalledProcessError as e:
raise RuntimeError(e.output.decode('utf-8'))
# vim:sw=4:ts=4:et: