stenotrainer/tts.py
2022-09-22 18:26:07 -06:00

150 lines
4.9 KiB
Python

# Copyright 2022 Mycroft AI Inc.
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU Affero General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU Affero General Public License for more details.
#
# You should have received a copy of the GNU Affero General Public License
# along with this program. If not, see <http://www.gnu.org/licenses/>.
#
import argparse
import os
import io
import logging
import threading
import tempfile
import typing
import simpleaudio as sa
import wave
from queue import Queue
from mimic3_tts import (
AudioResult,
Mimic3Settings,
Mimic3TextToSpeechSystem,
SSMLSpeaker,
)
from mimic3_tts.config import InferenceConfig
from mimic3_http.const import TextToWavParams
_LOGGER = logging.getLogger(__name__)
from dataclasses import dataclass
inf_conf = InferenceConfig()
def do_synthesis(params: TextToWavParams, mimic3: Mimic3TextToSpeechSystem) -> bytes:
"""Synthesize text into audio.
Returns: WAV bytes
"""
mimic3.speaker = None
mimic3.voice = params.voice
mimic3.settings.length_scale = params.length_scale
mimic3.settings.noise_scale = params.noise_scale
mimic3.settings.noise_w = params.noise_w
with io.BytesIO() as wav_io:
wav_file: wave.Wave_write = wave.open(wav_io, "wb")
wav_params_set = False
with wav_file:
try:
if params.ssml:
# SSML
results = SSMLSpeaker(mimic3).speak(params.text)
else:
# Plain text
mimic3.begin_utterance()
mimic3.speak_text(params.text, text_language=params.text_language)
results = mimic3.end_utterance()
for result in results:
# Add audio to existing WAV file
if isinstance(result, AudioResult):
if not wav_params_set:
wav_file.setframerate(result.sample_rate_hz)
wav_file.setsampwidth(result.sample_width_bytes)
wav_file.setnchannels(result.num_channels)
wav_params_set = True
wav_file.writeframes(result.audio_bytes)
except Exception as e:
if not wav_params_set:
# Set default parameters so exception can propagate
wav_file.setframerate(22050)
wav_file.setsampwidth(2)
wav_file.setnchannels(1)
raise e
wav_bytes = wav_io.getvalue()
return wav_bytes
def create_mimic3_system(
voice: str=None, speaker: str=None, length_scale:float=None,
rate: float = None,
noise_scale:float=None, noise_w:float=None, use_cuda: bool=None,
voices_directories: list[str]=None, use_deterministic_compute:bool=None,
preload_voices:list[str]=None
) -> Mimic3TextToSpeechSystem:
mimic3 = Mimic3TextToSpeechSystem(
Mimic3Settings(
rate=rate,
voice=voice,
speaker=speaker,
length_scale=length_scale,
noise_scale=noise_scale,
noise_w=noise_w,
use_cuda=use_cuda,
voices_directories=voices_directories,
use_deterministic_compute=use_deterministic_compute,
)
)
with mimic3:
if preload_voices:
# Ensure voices are preloaded
for voice_key in preload_voices:
mimic3.preload_voice(voice_key)
return mimic3
def play_wav_bytes(wav_bytes: bytes):
wav_file = tempfile.NamedTemporaryFile(mode="wb+", suffix=".wav", delete=False)
wav_file.write(wav_bytes)
wav_file.seek(0)
wav_file.close()
filename = wav_file.name
wave_obj = sa.WaveObject.from_wave_file(filename)
play_obj = wave_obj.play()
play_obj.wait_done()
os.unlink(filename)
def say_words(words: str, mimic3: Mimic3TextToSpeechSystem, length_scale: float = None, ssml=False):
"""Thread handler for synthesis requests"""
try:
with mimic3:
params = TextToWavParams(
text=words, voice=mimic3.settings.voice,
noise_scale=mimic3.settings.noise_scale,
noise_w=mimic3.settings.noise_w,
ssml=ssml,
length_scale=length_scale if length_scale is not None else mimic3.settings.length_scale
)
try:
play_wav_bytes(do_synthesis(params, mimic3))
except Exception as e:
raise(e)
except Exception as e:
raise(e)