import itertools import os import librosa import matplotlib.pyplot as plt import pyloudnorm import sounddevice import soundfile import torch from speechbrain.pretrained import EncoderClassifier from torchaudio.transforms import Resample from Modules.ToucanTTS.InferenceToucanTTS import ToucanTTS from Modules.Vocoder.HiFiGAN_Generator import HiFiGAN from Preprocessing.AudioPreprocessor import AudioPreprocessor from Preprocessing.TextFrontend import ArticulatoryCombinedTextFrontend from Preprocessing.TextFrontend import get_language_id from Utility.storage_config import MODELS_DIR from Utility.utils import cumsum_durations from Utility.utils import float2pcm class ToucanTTSInterface(torch.nn.Module): def __init__(self, device="cpu", # device that everything computes on. If a cuda device is available, this can speed things up by an order of magnitude. tts_model_path=os.path.join(MODELS_DIR, f"ToucanTTS_Meta", "best.pt"), # path to the ToucanTTS checkpoint or just a shorthand if run standalone vocoder_model_path=os.path.join(MODELS_DIR, f"Vocoder", "best.pt"), # path to the Vocoder checkpoint language="eng", # initial language of the model, can be changed later with the setter methods ): super().__init__() self.device = device if not tts_model_path.endswith(".pt"): # default to shorthand system tts_model_path = os.path.join(MODELS_DIR, f"ToucanTTS_{tts_model_path}", "best.pt") ################################ # build text to phone # ################################ self.text2phone = ArticulatoryCombinedTextFrontend(language=language, add_silence_to_end=True, device=device) ##################################### # load phone to features model # ##################################### checkpoint = torch.load(tts_model_path, map_location='cpu') self.phone2mel = ToucanTTS(weights=checkpoint["model"], config=checkpoint["config"]) with torch.no_grad(): self.phone2mel.store_inverse_all() # this also removes weight norm self.phone2mel = self.phone2mel.to(torch.device(device)) ###################################### # load features to style models # ###################################### self.speaker_embedding_func_ecapa = EncoderClassifier.from_hparams(source="speechbrain/spkrec-ecapa-voxceleb", run_opts={"device": str(device)}, savedir=os.path.join(MODELS_DIR, "Embedding", "speechbrain_speaker_embedding_ecapa")) ################################ # load mel to wave model # ################################ vocoder_checkpoint = torch.load(vocoder_model_path, map_location="cpu") self.vocoder = HiFiGAN() self.vocoder.load_state_dict(vocoder_checkpoint) self.vocoder = self.vocoder.to(device).eval() self.vocoder.remove_weight_norm() self.meter = pyloudnorm.Meter(24000) ################################ # set defaults # ################################ self.default_utterance_embedding = checkpoint["default_emb"].to(self.device) self.ap = AudioPreprocessor(input_sr=100, output_sr=16000, device=device) self.phone2mel.eval() self.vocoder.eval() self.lang_id = get_language_id(language) self.to(torch.device(device)) self.eval() def set_utterance_embedding(self, path_to_reference_audio="", embedding=None): if embedding is not None: self.default_utterance_embedding = embedding.squeeze().to(self.device) return if type(path_to_reference_audio) != list: path_to_reference_audio = [path_to_reference_audio] if len(path_to_reference_audio) > 0: for path in path_to_reference_audio: assert os.path.exists(path) speaker_embs = list() for path in path_to_reference_audio: wave, sr = soundfile.read(path) if len(wave.shape) > 1: # oh no, we found a stereo audio! if len(wave[0]) == 2: # let's figure out whether we need to switch the axes wave = wave.transpose() # if yes, we switch the axes. wave = librosa.to_mono(wave) wave = Resample(orig_freq=sr, new_freq=16000).to(self.device)(torch.tensor(wave, device=self.device, dtype=torch.float32)) speaker_embedding = self.speaker_embedding_func_ecapa.encode_batch(wavs=wave.to(self.device).squeeze().unsqueeze(0)).squeeze() speaker_embs.append(speaker_embedding) self.default_utterance_embedding = sum(speaker_embs) / len(speaker_embs) def set_language(self, lang_id): """ The id parameter actually refers to the shorthand. This has become ambiguous with the introduction of the actual language IDs """ self.set_phonemizer_language(lang_id=lang_id) self.set_accent_language(lang_id=lang_id) def set_phonemizer_language(self, lang_id): self.text2phone = ArticulatoryCombinedTextFrontend(language=lang_id, add_silence_to_end=True, device=self.device) def set_accent_language(self, lang_id): if lang_id in {'ajp', 'ajt', 'lak', 'lno', 'nul', 'pii', 'plj', 'slq', 'smd', 'snb', 'tpw', 'wya', 'zua', 'en-us', 'en-sc', 'fr-be', 'fr-sw', 'pt-br', 'spa-lat', 'vi-ctr', 'vi-so'}: if lang_id == 'vi-so' or lang_id == 'vi-ctr': lang_id = 'vie' elif lang_id == 'spa-lat': lang_id = 'spa' elif lang_id == 'pt-br': lang_id = 'por' elif lang_id == 'fr-sw' or lang_id == 'fr-be': lang_id = 'fra' elif lang_id == 'en-sc' or lang_id == 'en-us': lang_id = 'eng' else: # no clue where these others are even coming from, they are not in ISO 639-3 lang_id = 'eng' self.lang_id = get_language_id(lang_id).to(self.device) def forward(self, text, view=False, duration_scaling_factor=1.0, pitch_variance_scale=1.0, energy_variance_scale=1.0, pause_duration_scaling_factor=1.0, durations=None, pitch=None, energy=None, input_is_phones=False, return_plot_as_filepath=False, loudness_in_db=-24.0, prosody_creativity=0.1): """ duration_scaling_factor: reasonable values are 0.8 < scale < 1.2. 1.0 means no scaling happens, higher values increase durations for the whole utterance, lower values decrease durations for the whole utterance. pitch_variance_scale: reasonable values are 0.6 < scale < 1.4. 1.0 means no scaling happens, higher values increase variance of the pitch curve, lower values decrease variance of the pitch curve. energy_variance_scale: reasonable values are 0.6 < scale < 1.4. 1.0 means no scaling happens, higher values increase variance of the energy curve, lower values decrease variance of the energy curve. """ with torch.inference_mode(): phones = self.text2phone.string_to_tensor(text, input_phonemes=input_is_phones).to(torch.device(self.device)) mel, durations, pitch, energy = self.phone2mel(phones, return_duration_pitch_energy=True, utterance_embedding=self.default_utterance_embedding, durations=durations, pitch=pitch, energy=energy, lang_id=self.lang_id, duration_scaling_factor=duration_scaling_factor, pitch_variance_scale=pitch_variance_scale, energy_variance_scale=energy_variance_scale, pause_duration_scaling_factor=pause_duration_scaling_factor, prosody_creativity=prosody_creativity) wave, _, _ = self.vocoder(mel.unsqueeze(0)) wave = wave.squeeze().cpu() wave = wave.numpy() sr = 24000 try: loudness = self.meter.integrated_loudness(wave) wave = pyloudnorm.normalize.loudness(wave, loudness, loudness_in_db) except ValueError: # if the audio is too short, a value error will arise pass if view or return_plot_as_filepath: fig, ax = plt.subplots(nrows=1, ncols=1, figsize=(9, 5)) ax.imshow(mel.cpu().numpy(), origin="lower", cmap='GnBu') ax.yaxis.set_visible(False) duration_splits, label_positions = cumsum_durations(durations.cpu().numpy()) ax.xaxis.grid(True, which='minor') ax.set_xticks(label_positions, minor=False) if input_is_phones: phones = text.replace(" ", "|") else: phones = self.text2phone.get_phone_string(text, for_plot_labels=True) try: ax.set_xticklabels(phones) except IndexError: pass except ValueError: pass word_boundaries = list() for label_index, phone in enumerate(phones): if phone == "|": word_boundaries.append(label_positions[label_index]) try: prev_word_boundary = 0 word_label_positions = list() for word_boundary in word_boundaries: word_label_positions.append((word_boundary + prev_word_boundary) / 2) prev_word_boundary = word_boundary word_label_positions.append((duration_splits[-1] + prev_word_boundary) / 2) secondary_ax = ax.secondary_xaxis('bottom') secondary_ax.tick_params(axis="x", direction="out", pad=24) secondary_ax.set_xticks(word_label_positions, minor=False) secondary_ax.set_xticklabels(text.split()) secondary_ax.tick_params(axis='x', colors='orange') secondary_ax.xaxis.label.set_color('orange') except ValueError: ax.set_title(text) except IndexError: ax.set_title(text) ax.vlines(x=duration_splits, colors="green", linestyles="solid", ymin=0, ymax=120, linewidth=0.5) ax.vlines(x=word_boundaries, colors="orange", linestyles="solid", ymin=0, ymax=120, linewidth=1.0) plt.subplots_adjust(left=0.02, bottom=0.2, right=0.98, top=.9, wspace=0.0, hspace=0.0) ax.set_aspect("auto") if return_plot_as_filepath: plt.savefig("tmp.png") plt.close() return wave, sr, "tmp.png" return wave, sr def read_to_file(self, text_list, file_location, duration_scaling_factor=1.0, pitch_variance_scale=1.0, energy_variance_scale=1.0, pause_duration_scaling_factor=1.0, silent=False, dur_list=None, pitch_list=None, energy_list=None, prosody_creativity=0.1): """ Args: silent: Whether to be verbose about the process text_list: A list of strings to be read file_location: The path and name of the file it should be saved to energy_list: list of energy tensors to be used for the texts pitch_list: list of pitch tensors to be used for the texts dur_list: list of duration tensors to be used for the texts duration_scaling_factor: reasonable values are 0.8 < scale < 1.2. 1.0 means no scaling happens, higher values increase durations for the whole utterance, lower values decrease durations for the whole utterance. pause_duration_scaling_factor: reasonable values are 0.8 < scale < 1.2. 1.0 means no scaling happens, higher values increase durations for the pauses, lower values decrease durations for the whole utterance. pitch_variance_scale: reasonable values are 0.6 < scale < 1.4. 1.0 means no scaling happens, higher values increase variance of the pitch curve, lower values decrease variance of the pitch curve. energy_variance_scale: reasonable values are 0.6 < scale < 1.4. 1.0 means no scaling happens, higher values increase variance of the energy curve, lower values decrease variance of the energy curve. prosody_creativity: sampling temperature of the generative model that comes up with the pitch, energy and durations. Higher values mena more variance, lower temperature means less variance across generations. reasonable values are between 0.0 and 1.2, anything higher makes the voice sound very weird. """ if not dur_list: dur_list = [] if not pitch_list: pitch_list = [] if not energy_list: energy_list = [] silence = torch.zeros([400]) wav = silence.clone() for (text, durations, pitch, energy) in itertools.zip_longest(text_list, dur_list, pitch_list, energy_list): if text.strip() != "": if not silent: print("Now synthesizing: {}".format(text)) spoken_sentence, sr = self(text, durations=durations.to(self.device) if durations is not None else None, pitch=pitch.to(self.device) if pitch is not None else None, energy=energy.to(self.device) if energy is not None else None, duration_scaling_factor=duration_scaling_factor, pitch_variance_scale=pitch_variance_scale, energy_variance_scale=energy_variance_scale, pause_duration_scaling_factor=pause_duration_scaling_factor, prosody_creativity=prosody_creativity) spoken_sentence = torch.tensor(spoken_sentence).cpu() wav = torch.cat((wav, spoken_sentence, silence), 0) soundfile.write(file=file_location, data=float2pcm(wav), samplerate=sr, subtype="PCM_16") def read_aloud(self, text, view=False, duration_scaling_factor=1.0, pitch_variance_scale=1.0, energy_variance_scale=1.0, blocking=False, prosody_creativity=0.1): if text.strip() == "": return wav, sr = self(text, view, duration_scaling_factor=duration_scaling_factor, pitch_variance_scale=pitch_variance_scale, energy_variance_scale=energy_variance_scale, prosody_creativity=prosody_creativity) silence = torch.zeros([sr // 2]) wav = torch.cat((silence, torch.tensor(wav), silence), 0).numpy() sounddevice.play(float2pcm(wav), samplerate=sr) if view: plt.show() if blocking: sounddevice.wait()