diff --git a/mozilla_voice_tts/bin/train_tts.py b/mozilla_voice_tts/bin/train_tts.py index 1c27b5a4..25b80104 100644 --- a/mozilla_voice_tts/bin/train_tts.py +++ b/mozilla_voice_tts/bin/train_tts.py @@ -670,12 +670,12 @@ if __name__ == '__main__': args = parser.parse_args() if args.continue_path != '': + print(f" > Training continues for {args.continue_path}") args.output_path = args.continue_path args.config_path = os.path.join(args.continue_path, 'config.json') list_of_files = glob.glob(args.continue_path + "/*.pth.tar") # * means all if need specific format then *.csv latest_model_file = max(list_of_files, key=os.path.getctime) args.restore_path = latest_model_file - print(f" > Training continues for {args.restore_path}") # setup output paths and read configs c = load_config(args.config_path) diff --git a/mozilla_voice_tts/tts/datasets/preprocess.py b/mozilla_voice_tts/tts/datasets/preprocess.py index ece3bcb6..2389cd4c 100644 --- a/mozilla_voice_tts/tts/datasets/preprocess.py +++ b/mozilla_voice_tts/tts/datasets/preprocess.py @@ -251,4 +251,4 @@ def vctk(root_path, meta_files=None, wavs_path='wav48'): file_id + '.wav') items.append([text, wav_file, speaker_id]) - return items \ No newline at end of file + return items diff --git a/mozilla_voice_tts/utils/audio.py b/mozilla_voice_tts/utils/audio.py index 46c459f9..d81ba809 100644 --- a/mozilla_voice_tts/utils/audio.py +++ b/mozilla_voice_tts/utils/audio.py @@ -7,7 +7,7 @@ import pyworld as pw from mozilla_voice_tts.tts.utils.data import StandardScaler - +#pylint: disable=too-many-public-methods class AudioProcessor(object): def __init__(self, sample_rate=None, diff --git a/mozilla_voice_tts/utils/io.py b/mozilla_voice_tts/utils/io.py index 434c3a03..c96703ed 100644 --- a/mozilla_voice_tts/utils/io.py +++ b/mozilla_voice_tts/utils/io.py @@ -1,6 +1,5 @@ import re import json -from shutil import copyfile class AttrDict(dict): """A custom dict which converts dict keys diff --git a/tests/inputs/test_train_config.json b/tests/inputs/test_train_config.json index bea4cbb7..81a85729 100644 --- a/tests/inputs/test_train_config.json +++ b/tests/inputs/test_train_config.json @@ -1,161 +1,7 @@ -<<<<<<< HEAD:tests/inputs/test_train_config.json -{ - "model": "Tacotron2", - "run_name": "test_sample_dataset_run", - "run_description": "sample dataset test run", - - // AUDIO PARAMETERS - "audio":{ - // stft parameters - "fft_size": 1024, // number of stft frequency levels. Size of the linear spectogram frame. - "win_length": 1024, // stft window length in ms. - "hop_length": 256, // stft window hop-lengh in ms. - "frame_length_ms": null, // stft window length in ms.If null, 'win_length' is used. - "frame_shift_ms": null, // stft window hop-lengh in ms. If null, 'hop_length' is used. - - // Audio processing parameters - "sample_rate": 22050, // DATASET-RELATED: wav sample-rate. - "preemphasis": 0.0, // pre-emphasis to reduce spec noise and make it more structured. If 0.0, no -pre-emphasis. - "ref_level_db": 20, // reference level db, theoretically 20db is the sound of air. - - // Silence trimming - "do_trim_silence": true,// enable trimming of slience of audio as you load it. LJspeech (true), TWEB (false), Nancy (true) - "trim_db": 60, // threshold for timming silence. Set this according to your dataset. - - // Griffin-Lim - "power": 1.5, // value to sharpen wav signals after GL algorithm. - "griffin_lim_iters": 60,// #griffin-lim iterations. 30-60 is a good range. Larger the value, slower the generation. - - // MelSpectrogram parameters - "num_mels": 80, // size of the mel spec frame. - "mel_fmin": 0.0, // minimum freq level for mel-spec. ~50 for male and ~95 for female voices. Tune for dataset!! - "mel_fmax": 8000.0, // maximum freq level for mel-spec. Tune for dataset!! - "spec_gain": 20.0, - - // Normalization parameters - "signal_norm": true, // normalize spec values. Mean-Var normalization if 'stats_path' is defined otherwise range normalization defined by the other params. - "min_level_db": -100, // lower bound for normalization - "symmetric_norm": true, // move normalization to range [-1, 1] - "max_norm": 4.0, // scale normalization to range [-max_norm, max_norm] or [0, max_norm] - "clip_norm": true, // clip normalized values into the range. - "stats_path": null // DO NOT USE WITH MULTI_SPEAKER MODEL. scaler stats file computed by 'compute_statistics.py'. If it is defined, mean-std based notmalization is used and other normalization params are ignored - }, - - // VOCABULARY PARAMETERS - // if custom character set is not defined, - // default set in symbols.py is used - // "characters":{ - // "pad": "_", - // "eos": "~", - // "bos": "^", - // "characters": "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz!'(),-.:;? ", - // "punctuations":"!'(),-.:;? ", - // "phonemes":"iyɨʉɯuɪʏʊeøɘəɵɤoɛœɜɞʌɔæɐaɶɑɒᵻʘɓǀɗǃʄǂɠǁʛpbtdʈɖcɟkɡqɢʔɴŋɲɳnɱmʙrʀⱱɾɽɸβfvθðszʃʒʂʐçʝxɣχʁħʕhɦɬɮʋɹɻjɰlɭʎʟˈˌːˑʍwɥʜʢʡɕʑɺɧɚ˞ɫ" - // }, - - // DISTRIBUTED TRAINING - "distributed":{ - "backend": "nccl", - "url": "tcp:\/\/localhost:54321" - }, - - "reinit_layers": [], // give a list of layer names to restore from the given checkpoint. If not defined, it reloads all heuristically matching layers. - - // TRAINING - "batch_size": 1, // Batch size for training. Lower values than 32 might cause hard to learn attention. It is overwritten by 'gradual_training'. - "eval_batch_size":1, - "r": 7, // Number of decoder frames to predict per iteration. Set the initial values if gradual training is enabled. - "gradual_training": [[0, 7, 4]], //set gradual training steps [first_step, r, batch_size]. If it is null, gradual training is disabled. For Tacotron, you might need to reduce the 'batch_size' as you proceeed. - "loss_masking": true, // enable / disable loss masking against the sequence padding. - "ga_alpha": 10.0, // weight for guided attention loss. If > 0, guided attention is enabled. - "apex_amp_level": null, - - // VALIDATION - "run_eval": true, - "test_delay_epochs": 0, //Until attention is aligned, testing only wastes computation time. - "test_sentences_file": null, // set a file to load sentences to be used for testing. If it is null then we use default english sentences. - - // OPTIMIZER - "noam_schedule": false, // use noam warmup and lr schedule. - "grad_clip": 1.0, // upper limit for gradients for clipping. - "epochs": 1, // total number of epochs to train. - "lr": 0.0001, // Initial learning rate. If Noam decay is active, maximum learning rate. - "wd": 0.000001, // Weight decay weight. - "warmup_steps": 4000, // Noam decay steps to increase the learning rate from 0 to "lr" - "seq_len_norm": false, // Normalize eash sample loss with its length to alleviate imbalanced datasets. Use it if your dataset is small or has skewed distribution of sequence lengths. - - // TACOTRON PRENET - "memory_size": -1, // ONLY TACOTRON - size of the memory queue used fro storing last decoder predictions for auto-regression. If < 0, memory queue is disabled and decoder only uses the last prediction frame. - "prenet_type": "bn", // "original" or "bn". - "prenet_dropout": false, // enable/disable dropout at prenet. - - // TACOTRON ATTENTION - "attention_type": "original", // 'original' or 'graves' - "attention_heads": 4, // number of attention heads (only for 'graves') - "attention_norm": "sigmoid", // softmax or sigmoid. - "windowing": false, // Enables attention windowing. Used only in eval mode. - "use_forward_attn": false, // if it uses forward attention. In general, it aligns faster. - "forward_attn_mask": false, // Additional masking forcing monotonicity only in eval mode. - "transition_agent": false, // enable/disable transition agent of forward attention. - "location_attn": true, // enable_disable location sensitive attention. It is enabled for TACOTRON by default. - "bidirectional_decoder": false, // use https://arxiv.org/abs/1907.09006. Use it, if attention does not work well with your dataset. - "double_decoder_consistency": true, // use DDC explained here https://erogol.com/solving-attention-problems-of-tts-models-with-double-decoder-consistency-draft/ - "ddc_r": 7, // reduction rate for coarse decoder. - - // STOPNET - "stopnet": true, // Train stopnet predicting the end of synthesis. - "separate_stopnet": true, // Train stopnet seperately if 'stopnet==true'. It prevents stopnet loss to influence the rest of the model. It causes a better model, but it trains SLOWER. - - // TENSORBOARD and LOGGING - "print_step": 1, // Number of steps to log training on console. - "tb_plot_step": 100, // Number of steps to plot TB training figures. - "print_eval": false, // If True, it prints intermediate loss values in evalulation. - "save_step": 10000, // Number of training steps expected to save traninpg stats and checkpoints. - "checkpoint": true, // If true, it saves checkpoints per "save_step" - "tb_model_param_stats": false, // true, plots param stats per layer on tensorboard. Might be memory consuming, but good for debugging. - - // DATA LOADING - "text_cleaner": "phoneme_cleaners", - "enable_eos_bos_chars": false, // enable/disable beginning of sentence and end of sentence chars. - "num_loader_workers": 4, // number of training data loader processes. Don't set it too big. 4-8 are good values. - "num_val_loader_workers": 4, // number of evaluation data loader processes. - "batch_group_size": 0, //Number of batches to shuffle after bucketing. - "min_seq_len": 6, // DATASET-RELATED: minimum text length to use in training - "max_seq_len": 153, // DATASET-RELATED: maximum text length - - // PATHS - "output_path": "tests/train_outputs/", - - // PHONEMES - "phoneme_cache_path": "tests/train_outputs/phoneme_cache/", // phoneme computation is slow, therefore, it caches results in the given folder. - "use_phonemes": true, // use phonemes instead of raw characters. It is suggested for better pronounciation. - "phoneme_language": "en-us", // depending on your target language, pick one from https://github.com/bootphon/phonemizer#languages - - // MULTI-SPEAKER and GST - "use_speaker_embedding": false, // use speaker embedding to enable multi-speaker learning. - "style_wav_for_test": null, // path to style wav file to be used in TacotronGST inference. - "use_gst": false, // TACOTRON ONLY: use global style tokens - - // DATASETS - "train_portion": 0.1, // dataset portion used for training. It is mainly for internal experiments. - "eval_portion": 0.1, // dataset portion used for training. It is mainly for internal experiments. - "datasets": // List of datasets. They all merged and they get different speaker_ids. - [ - { - "name": "ljspeech", - "path": "tests/data/ljspeech/", - "meta_file_train": "metadata.csv", - "meta_file_val": "metadata.csv" - } - ] - -} - -======= { "model": "Tacotron2", - "run_name": "ljspeech-ddc-bn", - "run_description": "tacotron2 with ddc and batch-normalization", + "run_name": "test_sample_dataset_run", + "run_description": "sample dataset test run", // AUDIO PARAMETERS "audio":{ @@ -183,7 +29,7 @@ "num_mels": 80, // size of the mel spec frame. "mel_fmin": 0.0, // minimum freq level for mel-spec. ~50 for male and ~95 for female voices. Tune for dataset!! "mel_fmax": 8000.0, // maximum freq level for mel-spec. Tune for dataset!! - "spec_gain": 20, + "spec_gain": 20.0, // Normalization parameters "signal_norm": true, // normalize spec values. Mean-Var normalization if 'stats_path' is defined otherwise range normalization defined by the other params. @@ -215,30 +61,31 @@ "reinit_layers": [], // give a list of layer names to restore from the given checkpoint. If not defined, it reloads all heuristically matching layers. // TRAINING - "batch_size": 32, // Batch size for training. Lower values than 32 might cause hard to learn attention. It is overwritten by 'gradual_training'. - "eval_batch_size":16, + "batch_size": 1, // Batch size for training. Lower values than 32 might cause hard to learn attention. It is overwritten by 'gradual_training'. + "eval_batch_size":1, "r": 7, // Number of decoder frames to predict per iteration. Set the initial values if gradual training is enabled. - "gradual_training": [[0, 7, 64], [1, 5, 64], [50000, 3, 32], [130000, 2, 32], [290000, 1, 32]], //set gradual training steps [first_step, r, batch_size]. If it is null, gradual training is disabled. For Tacotron, you might need to reduce the 'batch_size' as you proceeed. + "gradual_training": [[0, 7, 4]], //set gradual training steps [first_step, r, batch_size]. If it is null, gradual training is disabled. For Tacotron, you might need to reduce the 'batch_size' as you proceeed. "loss_masking": true, // enable / disable loss masking against the sequence padding. "ga_alpha": 10.0, // weight for guided attention loss. If > 0, guided attention is enabled. + "apex_amp_level": null, // VALIDATION "run_eval": true, - "test_delay_epochs": 10, //Until attention is aligned, testing only wastes computation time. + "test_delay_epochs": 0, //Until attention is aligned, testing only wastes computation time. "test_sentences_file": null, // set a file to load sentences to be used for testing. If it is null then we use default english sentences. // OPTIMIZER "noam_schedule": false, // use noam warmup and lr schedule. "grad_clip": 1.0, // upper limit for gradients for clipping. - "epochs": 1000, // total number of epochs to train. + "epochs": 1, // total number of epochs to train. "lr": 0.0001, // Initial learning rate. If Noam decay is active, maximum learning rate. "wd": 0.000001, // Weight decay weight. "warmup_steps": 4000, // Noam decay steps to increase the learning rate from 0 to "lr" "seq_len_norm": false, // Normalize eash sample loss with its length to alleviate imbalanced datasets. Use it if your dataset is small or has skewed distribution of sequence lengths. // TACOTRON PRENET - "memory_size": -1, // ONLY TACOTRON - size of the memory queue used fro storing last decoder predictions for auto-regression. If < 0, memory queue is disabled and decoder only uses the last prediction frame. - "prenet_type": "bn", // "original" or "bn". + "memory_size": -1, // ONLY TACOTRON - size of the memory queue used fro storing last decoder predictions for auto-regression. If < 0, memory queue is disabled and decoder only uses the last prediction frame. + "prenet_type": "bn", // "original" or "bn". "prenet_dropout": false, // enable/disable dropout at prenet. // TACOTRON ATTENTION @@ -259,8 +106,8 @@ "separate_stopnet": true, // Train stopnet seperately if 'stopnet==true'. It prevents stopnet loss to influence the rest of the model. It causes a better model, but it trains SLOWER. // TENSORBOARD and LOGGING - "print_step": 25, // Number of steps to log training on console. - "tb_plot_step:": 100, // Number of steps to plot TB training figures. + "print_step": 1, // Number of steps to log training on console. + "tb_plot_step": 100, // Number of steps to plot TB training figures. "print_eval": false, // If True, it prints intermediate loss values in evalulation. "save_step": 10000, // Number of training steps expected to save traninpg stats and checkpoints. "checkpoint": true, // If true, it saves checkpoints per "save_step" @@ -276,36 +123,40 @@ "max_seq_len": 153, // DATASET-RELATED: maximum text length // PATHS - "output_path": "/home/erogol/Models/LJSpeech/", + "output_path": "tests/train_outputs/", // PHONEMES - "phoneme_cache_path": "/media/erogol/data_ssd2/mozilla_us_phonemes_3", // phoneme computation is slow, therefore, it caches results in the given folder. + "phoneme_cache_path": "tests/train_outputs/phoneme_cache/", // phoneme computation is slow, therefore, it caches results in the given folder. "use_phonemes": true, // use phonemes instead of raw characters. It is suggested for better pronounciation. "phoneme_language": "en-us", // depending on your target language, pick one from https://github.com/bootphon/phonemizer#languages // MULTI-SPEAKER and GST + "use_external_speaker_embedding_file": false, + "external_speaker_embedding_file": null, "use_speaker_embedding": false, // use speaker embedding to enable multi-speaker learning. "use_gst": true, // use global style tokens "gst": { // gst parameter if gst is enabled - "gst_style_input": null, // Condition the style input either on a - // -> wave file [path to wave] or - // -> dictionary using the style tokens {'token1': 'value', 'token2': 'value'} example {"0": 0.15, "1": 0.15, "5": -0.15} + "gst_style_input": null, // Condition the style input either on a + // -> wave file [path to wave] or + // -> dictionary using the style tokens {'token1': 'value', 'token2': 'value'} example {"0": 0.15, "1": 0.15, "5": -0.15} // with the dictionary being len(dict) == len(gst_style_tokens). - "gst_embedding_dim": 512, + "gst_embedding_dim": 512, "gst_num_heads": 4, "gst_style_tokens": 10 - }, + }, // DATASETS + "train_portion": 0.1, // dataset portion used for training. It is mainly for internal experiments. + "eval_portion": 0.1, // dataset portion used for training. It is mainly for internal experiments. "datasets": // List of datasets. They all merged and they get different speaker_ids. [ { "name": "ljspeech", - "path": "/home/erogol/Data/LJSpeech-1.1/", + "path": "tests/data/ljspeech/", "meta_file_train": "metadata.csv", - "meta_file_val": null + "meta_file_val": "metadata.csv" } ] + } ->>>>>>> Added support for Tacotron2 GST + abbility to condition style input with wav or tokens:config.json diff --git a/tests/inputs/test_vocoder_multiband_melgan_config.json b/tests/inputs/test_vocoder_multiband_melgan_config.json index c0f552a4..442550c6 100644 --- a/tests/inputs/test_vocoder_multiband_melgan_config.json +++ b/tests/inputs/test_vocoder_multiband_melgan_config.json @@ -139,6 +139,6 @@ "eval_split_size": 10, // PATHS - "output_path": "tests/outputs/train_outputs/" + "output_path": "tests/train_outputs/" } diff --git a/tests/test_vocoder_train.sh b/tests/test_vocoder_train.sh index 6be7177d..bba730dd 100755 --- a/tests/test_vocoder_train.sh +++ b/tests/test_vocoder_train.sh @@ -7,9 +7,9 @@ mkdir $BASEDIR/train_outputs # run training CUDA_VISIBLE_DEVICES="" python mozilla_voice_tts/bin/train_vocoder.py --config_path $BASEDIR/inputs/test_vocoder_multiband_melgan_config.json # find the training folder -LATEST_FOLDER=$(ls $BASEDIR/outputs/train_outputs/| sort | tail -1) +LATEST_FOLDER=$(ls $BASEDIR/train_outputs/| sort | tail -1) echo $LATEST_FOLDER # continue the previous training -CUDA_VISIBLE_DEVICES="" python mozilla_voice_tts/bin/train_vocoder.py --continue_path $BASEDIR/outputs/train_outputs/$LATEST_FOLDER +CUDA_VISIBLE_DEVICES="" python mozilla_voice_tts/bin/train_vocoder.py --continue_path $BASEDIR/train_outputs/$LATEST_FOLDER # remove all the outputs -rm -rf $BASEDIR/train_outputs/ +rm -rf $BASEDIR/train_outputs/$LATEST_FOLDER