o j/%@sxddlmZmZddlmZmZddlmZddlm Z eGdddeZ eGdddeZ eGd d d e Z d S) ) dataclassfield)ListOptional)Coqpit) BaseVCConfigc@seZdZUdZeddZeed<eddZe ed<eddZ e ed<ed dZ e ed <ed dZ e ed <ed dZ e ed <eddZe ed<eddZeed<eddZeeed<dS)FreeVCAudioConfigaAudio configuration Args: max_wav_value (float): The maximum value of the waveform. input_sample_rate (int): The sampling rate of the input waveform. output_sample_rate (int): The sampling rate of the output waveform. filter_length (int): The length of the filter. hop_length (int): The hop length. win_length (int): The window length. n_mel_channels (int): The number of mel channels. mel_fmin (float): The minimum frequency of the mel filterbank. mel_fmax (Optional[float]): The maximum frequency of the mel filterbank. g@default max_wav_valuei>input_sample_ratei]output_sample_ratei filter_lengthi@ hop_length win_lengthPn_mel_channelsgmel_fminNmel_fmax)__name__ __module__ __qualname____doc__rr float__annotations__r intr rrrrrrrrrO/home/kuhnn/.local/lib/python3.10/site-packages/TTS/vc/configs/freevc_config.pyr s rc@seZdZUdZeddZeed<eddZeed<eddZ eed<eddZ eed <ed dZ eed <ed dZ eed <eddZ eed<eddZeed<eddZeed<edddZeeed<edddZeeeed<edddZeeed<eddZeed<edddZeeed<eddZeed <ed!dZeed"<ed#dZeed$<ed%dZeed&<ed!dZeed'<ed(dZeed)<ed*dZeed+<d,S)- FreeVCArgsagFreeVC model arguments Args: spec_channels (int): The number of channels in the spectrogram. inter_channels (int): The number of channels in the intermediate layers. hidden_channels (int): The number of channels in the hidden layers. filter_channels (int): The number of channels in the filter layers. n_heads (int): The number of attention heads. n_layers (int): The number of layers. kernel_size (int): The size of the kernel. p_dropout (float): The dropout probability. resblock (str): The type of residual block. resblock_kernel_sizes (List[int]): The kernel sizes for the residual blocks. resblock_dilation_sizes (List[List[int]]): The dilation sizes for the residual blocks. upsample_rates (List[int]): The upsample rates. upsample_initial_channel (int): The number of channels in the initial upsample layer. upsample_kernel_sizes (List[int]): The kernel sizes for the upsample layers. n_layers_q (int): The number of layers in the quantization network. use_spectral_norm (bool): Whether to use spectral normalization. gin_channels (int): The number of channels in the global conditioning vector. ssl_dim (int): The dimension of the self-supervised learning embedding. use_spk (bool): Whether to use external speaker encoder. ir spec_channelsinter_channelshidden_channelsifilter_channelsn_headsn_layers kernel_sizeg? p_dropout1resblockcCgdS)N)r( rrrrr}zFreeVCArgs.default_factoryresblock_kernel_sizescCsgdgdgdgS)N)r(rrrrrr0~sresblock_dilation_sizescCr-)N) r$r$rrrrrr0r1upsample_ratesiupsample_initial_channelcCr-)N)r<r=rrrrrr0r1upsample_kernel_sizes n_layers_qFuse_spectral_norm gin_channelsissl_dimuse_spkrnum_spksi# segment_sizeN) rrrrrrrrr!r"r#r%r'r)r*rr,strr4rr7r:r;r>r?r@boolrBrCrDrErFrrrrr5s. =rc@seZdZUdZdZeed<eedZ eed<ee dZ e ed<dZ e ed<dZe ed <d Ze ed <ed d dZeed<edd dZeed<dZeed<dZe ed<dZeed<dZeed<dZeed<d Ze ed<dZeeed<dZeed<ddZdS) FreeVCConfigasDefines parameters for FreeVC End2End TTS model. Args: model (str): Model name. Do not change unless you know what you are doing. model_args (FreeVCArgs): Model architecture arguments. Defaults to `FreeVCArgs()`. audio (FreeVCAudioConfig): Audio processing configuration. Defaults to `FreeVCAudioConfig()`. grad_clip (List): Gradient clipping thresholds for each optimizer. Defaults to `[1000.0, 1000.0]`. lr_gen (float): Initial learning rate for the generator. Defaults to 0.0002. lr_disc (float): Initial learning rate for the discriminator. Defaults to 0.0002. lr_scheduler_gen (str): Name of the learning rate scheduler for the generator. One of the `torch.optim.lr_scheduler.*`. Defaults to `ExponentialLR`. lr_scheduler_gen_params (dict): Parameters for the learning rate scheduler of the generator. Defaults to `{'gamma': 0.999875, "last_epoch":-1}`. lr_scheduler_disc (str): Name of the learning rate scheduler for the discriminator. One of the `torch.optim.lr_scheduler.*`. Defaults to `ExponentialLR`. lr_scheduler_disc_params (dict): Parameters for the learning rate scheduler of the discriminator. Defaults to `{'gamma': 0.999875, "last_epoch":-1}`. scheduler_after_epoch (bool): If true, step the schedulers after each epoch else after each step. Defaults to `False`. optimizer (str): Name of the optimizer to use with both the generator and the discriminator networks. One of the `torch.optim.*`. Defaults to `AdamW`. kl_loss_alpha (float): Loss weight for KL loss. Defaults to 1.0. disc_loss_alpha (float): Loss weight for the discriminator loss. Defaults to 1.0. gen_loss_alpha (float): Loss weight for the generator loss. Defaults to 1.0. feat_loss_alpha (float): Loss weight for the feature matching loss. Defaults to 1.0. mel_loss_alpha (float): Loss weight for the mel loss. Defaults to 45.0. return_wav (bool): If true, data loader returns the waveform as well as the other outputs. Do not change. Defaults to `True`. compute_linear_spec (bool): If true, the linear spectrogram is computed and returned alongside the mel output. Do not change. Defaults to `True`. use_weighted_sampler (bool): If true, use weighted sampler with bucketing for balancing samples between datasets used in training. Defaults to `False`. weighted_sampler_attrs (dict): Key retuned by the formatter to be used for weighted sampler. For example `{"root_path": 2.0, "speaker_name": 1.0}` sets sample probabilities by overweighting `root_path` by 2.0. Defaults to `{}`. weighted_sampler_multipliers (dict): Weight each unique value of a key returned by the formatter for weighted sampling. For example `{"root_path":{"/raid/datasets/libritts-clean-16khz-bwe-coqui_44khz/LibriTTS/train-clean-100/":1.0, "/raid/datasets/libritts-clean-16khz-bwe-coqui_44khz/LibriTTS/train-clean-360/": 0.5}`. It will sample instances from `train-clean-100` 2 times more than `train-clean-360`. Defaults to `{}`. r (int): Number of spectrogram frames to be generated at a time. Do not change. Defaults to `1`. add_blank (bool): If true, a blank token is added in between every character. Defaults to `True`. test_sentences (List[List]): List of sentences with speaker and language information to be used for testing. language_ids_file (str): Path to the language ids file. use_language_embedding (bool): If true, language embedding is used. Defaults to `False`. Note: Check :class:`TTS.tts.configs.shared_configs.BaseTTSConfig` for the inherited parameters. Example: >>> from TTS.vc.configs.freevc_config import FreeVCConfig >>> config = FreeVCConfig() freevcmodelr2 model_argsaudioT return_wavcompute_linear_specFuse_weighted_samplercCiSNrrrrrr0zFreeVCConfig.weighted_sampler_attrscCrQrRrrrrrr0rSweighted_sampler_multipliersr5r add_blankr num_speakersN speakers_filerAspeaker_embedding_channelsuse_d_vector_file d_vector_file d_vector_dimcCs*|jD] \}}t||r|||<qdSrR)rLitemshasattr)selfkeyvalrrr __post_init__s  zFreeVCConfig.__post_init__)rrrrrKrGrrrrLrrMrNrHrOrPrTdictrUrVrrWrXrYrZr[r\rr]rcrrrrrIs&  c          rIN) dataclassesrrtypingrrcoqpitrTTS.vc.configs.shared_configsrrrrIrrrrs  +U