MockingBird/toolbox/__init__.py
Vega 2a99f0ff05
Add gst (#137)
* Commit with working GST

* Make it backward compatible

* Add readme
2021-10-12 19:43:29 +08:00

379 lines
14 KiB
Python

from toolbox.ui import UI
from encoder import inference as encoder
from synthesizer.inference import Synthesizer
from vocoder.wavernn import inference as rnn_vocoder
from vocoder.hifigan import inference as gan_vocoder
from pathlib import Path
from time import perf_counter as timer
from toolbox.utterance import Utterance
import numpy as np
import traceback
import sys
import torch
import librosa
import re
from audioread.exceptions import NoBackendError
# 默认使用wavernn
vocoder = rnn_vocoder
# Use this directory structure for your datasets, or modify it to fit your needs
recognized_datasets = [
"LibriSpeech/dev-clean",
"LibriSpeech/dev-other",
"LibriSpeech/test-clean",
"LibriSpeech/test-other",
"LibriSpeech/train-clean-100",
"LibriSpeech/train-clean-360",
"LibriSpeech/train-other-500",
"LibriTTS/dev-clean",
"LibriTTS/dev-other",
"LibriTTS/test-clean",
"LibriTTS/test-other",
"LibriTTS/train-clean-100",
"LibriTTS/train-clean-360",
"LibriTTS/train-other-500",
"LJSpeech-1.1",
"VoxCeleb1/wav",
"VoxCeleb1/test_wav",
"VoxCeleb2/dev/aac",
"VoxCeleb2/test/aac",
"VCTK-Corpus/wav48",
"aidatatang_200zh/corpus/dev",
"aidatatang_200zh/corpus/test",
"aishell3/test/wav",
"magicdata/train",
]
#Maximum of generated wavs to keep on memory
MAX_WAVES = 15
class Toolbox:
def __init__(self, datasets_root, enc_models_dir, syn_models_dir, voc_models_dir, seed, no_mp3_support):
self.no_mp3_support = no_mp3_support
sys.excepthook = self.excepthook
self.datasets_root = datasets_root
self.utterances = set()
self.current_generated = (None, None, None, None) # speaker_name, spec, breaks, wav
self.synthesizer = None # type: Synthesizer
self.current_wav = None
self.waves_list = []
self.waves_count = 0
self.waves_namelist = []
# Check for webrtcvad (enables removal of silences in vocoder output)
try:
import webrtcvad
self.trim_silences = True
except:
self.trim_silences = False
# Initialize the events and the interface
self.ui = UI()
self.style_idx = 0
self.reset_ui(enc_models_dir, syn_models_dir, voc_models_dir, seed)
self.setup_events()
self.ui.start()
def excepthook(self, exc_type, exc_value, exc_tb):
traceback.print_exception(exc_type, exc_value, exc_tb)
self.ui.log("Exception: %s" % exc_value)
def setup_events(self):
# Dataset, speaker and utterance selection
self.ui.browser_load_button.clicked.connect(lambda: self.load_from_browser())
random_func = lambda level: lambda: self.ui.populate_browser(self.datasets_root,
recognized_datasets,
level)
self.ui.random_dataset_button.clicked.connect(random_func(0))
self.ui.random_speaker_button.clicked.connect(random_func(1))
self.ui.random_utterance_button.clicked.connect(random_func(2))
self.ui.dataset_box.currentIndexChanged.connect(random_func(1))
self.ui.speaker_box.currentIndexChanged.connect(random_func(2))
# Model selection
self.ui.encoder_box.currentIndexChanged.connect(self.init_encoder)
def func():
self.synthesizer = None
self.ui.synthesizer_box.currentIndexChanged.connect(func)
self.ui.vocoder_box.currentIndexChanged.connect(self.init_vocoder)
# Utterance selection
func = lambda: self.load_from_browser(self.ui.browse_file())
self.ui.browser_browse_button.clicked.connect(func)
func = lambda: self.ui.draw_utterance(self.ui.selected_utterance, "current")
self.ui.utterance_history.currentIndexChanged.connect(func)
func = lambda: self.ui.play(self.ui.selected_utterance.wav, Synthesizer.sample_rate)
self.ui.play_button.clicked.connect(func)
self.ui.stop_button.clicked.connect(self.ui.stop)
self.ui.record_button.clicked.connect(self.record)
#Audio
self.ui.setup_audio_devices(Synthesizer.sample_rate)
#Wav playback & save
func = lambda: self.replay_last_wav()
self.ui.replay_wav_button.clicked.connect(func)
func = lambda: self.export_current_wave()
self.ui.export_wav_button.clicked.connect(func)
self.ui.waves_cb.currentIndexChanged.connect(self.set_current_wav)
# Generation
func = lambda: self.synthesize() or self.vocode()
self.ui.generate_button.clicked.connect(func)
self.ui.synthesize_button.clicked.connect(self.synthesize)
self.ui.vocode_button.clicked.connect(self.vocode)
self.ui.random_seed_checkbox.clicked.connect(self.update_seed_textbox)
# UMAP legend
self.ui.clear_button.clicked.connect(self.clear_utterances)
def set_current_wav(self, index):
self.current_wav = self.waves_list[index]
def export_current_wave(self):
self.ui.save_audio_file(self.current_wav, Synthesizer.sample_rate)
def replay_last_wav(self):
self.ui.play(self.current_wav, Synthesizer.sample_rate)
def reset_ui(self, encoder_models_dir, synthesizer_models_dir, vocoder_models_dir, seed):
self.ui.populate_browser(self.datasets_root, recognized_datasets, 0, True)
self.ui.populate_models(encoder_models_dir, synthesizer_models_dir, vocoder_models_dir)
self.ui.populate_gen_options(seed, self.trim_silences)
def load_from_browser(self, fpath=None):
if fpath is None:
fpath = Path(self.datasets_root,
self.ui.current_dataset_name,
self.ui.current_speaker_name,
self.ui.current_utterance_name)
name = str(fpath.relative_to(self.datasets_root))
speaker_name = self.ui.current_dataset_name + '_' + self.ui.current_speaker_name
# Select the next utterance
if self.ui.auto_next_checkbox.isChecked():
self.ui.browser_select_next()
elif fpath == "":
return
else:
name = fpath.name
speaker_name = fpath.parent.name
if fpath.suffix.lower() == ".mp3" and self.no_mp3_support:
self.ui.log("Error: No mp3 file argument was passed but an mp3 file was used")
return
# Get the wav from the disk. We take the wav with the vocoder/synthesizer format for
# playback, so as to have a fair comparison with the generated audio
wav = Synthesizer.load_preprocess_wav(fpath)
self.ui.log("Loaded %s" % name)
self.add_real_utterance(wav, name, speaker_name)
def record(self):
wav = self.ui.record_one(encoder.sampling_rate, 5)
if wav is None:
return
self.ui.play(wav, encoder.sampling_rate)
speaker_name = "user01"
name = speaker_name + "_rec_%05d" % np.random.randint(100000)
self.add_real_utterance(wav, name, speaker_name)
def add_real_utterance(self, wav, name, speaker_name):
# Compute the mel spectrogram
spec = Synthesizer.make_spectrogram(wav)
self.ui.draw_spec(spec, "current")
# Compute the embedding
if not encoder.is_loaded():
self.init_encoder()
encoder_wav = encoder.preprocess_wav(wav)
embed, partial_embeds, _ = encoder.embed_utterance(encoder_wav, return_partials=True)
# Add the utterance
utterance = Utterance(name, speaker_name, wav, spec, embed, partial_embeds, False)
self.utterances.add(utterance)
self.ui.register_utterance(utterance)
# Plot it
self.ui.draw_embed(embed, name, "current")
self.ui.draw_umap_projections(self.utterances)
def clear_utterances(self):
self.utterances.clear()
self.ui.draw_umap_projections(self.utterances)
def synthesize(self):
self.ui.log("Generating the mel spectrogram...")
self.ui.set_loading(1)
# Update the synthesizer random seed
if self.ui.random_seed_checkbox.isChecked():
seed = int(self.ui.seed_textbox.text())
self.ui.populate_gen_options(seed, self.trim_silences)
else:
seed = None
if seed is not None:
torch.manual_seed(seed)
# Synthesize the spectrogram
if self.synthesizer is None or seed is not None:
self.init_synthesizer()
texts = self.ui.text_prompt.toPlainText().split("\n")
punctuation = '!,。、,' # punctuate and split/clean text
processed_texts = []
for text in texts:
for processed_text in re.sub(r'[{}]+'.format(punctuation), '\n', text).split('\n'):
if processed_text:
processed_texts.append(processed_text.strip())
texts = processed_texts
embed = self.ui.selected_utterance.embed
embeds = [embed] * len(texts)
specs = self.synthesizer.synthesize_spectrograms(texts, embeds, style_idx=int(self.ui.style_idx_textbox.text()))
breaks = [spec.shape[1] for spec in specs]
spec = np.concatenate(specs, axis=1)
self.ui.draw_spec(spec, "generated")
self.current_generated = (self.ui.selected_utterance.speaker_name, spec, breaks, None)
self.ui.set_loading(0)
def vocode(self):
speaker_name, spec, breaks, _ = self.current_generated
assert spec is not None
# Initialize the vocoder model and make it determinstic, if user provides a seed
if self.ui.random_seed_checkbox.isChecked():
seed = int(self.ui.seed_textbox.text())
self.ui.populate_gen_options(seed, self.trim_silences)
else:
seed = None
if seed is not None:
torch.manual_seed(seed)
# Synthesize the waveform
if not vocoder.is_loaded() or seed is not None:
self.init_vocoder()
def vocoder_progress(i, seq_len, b_size, gen_rate):
real_time_factor = (gen_rate / Synthesizer.sample_rate) * 1000
line = "Waveform generation: %d/%d (batch size: %d, rate: %.1fkHz - %.2fx real time)" \
% (i * b_size, seq_len * b_size, b_size, gen_rate, real_time_factor)
self.ui.log(line, "overwrite")
self.ui.set_loading(i, seq_len)
if self.ui.current_vocoder_fpath is not None:
self.ui.log("")
wav = vocoder.infer_waveform(spec, progress_callback=vocoder_progress)
else:
self.ui.log("Waveform generation with Griffin-Lim... ")
wav = Synthesizer.griffin_lim(spec)
self.ui.set_loading(0)
self.ui.log(" Done!", "append")
# Add breaks
b_ends = np.cumsum(np.array(breaks) * Synthesizer.hparams.hop_size)
b_starts = np.concatenate(([0], b_ends[:-1]))
wavs = [wav[start:end] for start, end, in zip(b_starts, b_ends)]
breaks = [np.zeros(int(0.15 * Synthesizer.sample_rate))] * len(breaks)
wav = np.concatenate([i for w, b in zip(wavs, breaks) for i in (w, b)])
# Trim excessive silences
if self.ui.trim_silences_checkbox.isChecked():
wav = encoder.preprocess_wav(wav)
# Play it
wav = wav / np.abs(wav).max() * 0.97
self.ui.play(wav, Synthesizer.sample_rate)
# Name it (history displayed in combobox)
# TODO better naming for the combobox items?
wav_name = str(self.waves_count + 1)
#Update waves combobox
self.waves_count += 1
if self.waves_count > MAX_WAVES:
self.waves_list.pop()
self.waves_namelist.pop()
self.waves_list.insert(0, wav)
self.waves_namelist.insert(0, wav_name)
self.ui.waves_cb.disconnect()
self.ui.waves_cb_model.setStringList(self.waves_namelist)
self.ui.waves_cb.setCurrentIndex(0)
self.ui.waves_cb.currentIndexChanged.connect(self.set_current_wav)
# Update current wav
self.set_current_wav(0)
#Enable replay and save buttons:
self.ui.replay_wav_button.setDisabled(False)
self.ui.export_wav_button.setDisabled(False)
# Compute the embedding
# TODO: this is problematic with different sampling rates, gotta fix it
if not encoder.is_loaded():
self.init_encoder()
encoder_wav = encoder.preprocess_wav(wav)
embed, partial_embeds, _ = encoder.embed_utterance(encoder_wav, return_partials=True)
# Add the utterance
name = speaker_name + "_gen_%05d" % np.random.randint(100000)
utterance = Utterance(name, speaker_name, wav, spec, embed, partial_embeds, True)
self.utterances.add(utterance)
# Plot it
self.ui.draw_embed(embed, name, "generated")
self.ui.draw_umap_projections(self.utterances)
def init_encoder(self):
model_fpath = self.ui.current_encoder_fpath
self.ui.log("Loading the encoder %s... " % model_fpath)
self.ui.set_loading(1)
start = timer()
encoder.load_model(model_fpath)
self.ui.log("Done (%dms)." % int(1000 * (timer() - start)), "append")
self.ui.set_loading(0)
def init_synthesizer(self):
model_fpath = self.ui.current_synthesizer_fpath
self.ui.log("Loading the synthesizer %s... " % model_fpath)
self.ui.set_loading(1)
start = timer()
self.synthesizer = Synthesizer(model_fpath)
self.ui.log("Done (%dms)." % int(1000 * (timer() - start)), "append")
self.ui.set_loading(0)
def init_vocoder(self):
global vocoder
model_fpath = self.ui.current_vocoder_fpath
# Case of Griffin-lim
if model_fpath is None:
return
# Sekect vocoder based on model name
if model_fpath.name[0] == "g":
vocoder = gan_vocoder
self.ui.log("set hifigan as vocoder")
else:
vocoder = rnn_vocoder
self.ui.log("set wavernn as vocoder")
self.ui.log("Loading the vocoder %s... " % model_fpath)
self.ui.set_loading(1)
start = timer()
vocoder.load_model(model_fpath)
self.ui.log("Done (%dms)." % int(1000 * (timer() - start)), "append")
self.ui.set_loading(0)
def update_seed_textbox(self):
self.ui.update_seed_textbox()