Upload 95 files

.gitattributes

@@ -33,3 +33,7 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+vocoder/female/aryan/hifigan/generator filter=lfs diff=lfs merge=lfs -text
+vocoder/female/dravidian/hifigan/generator filter=lfs diff=lfs merge=lfs -text
+vocoder/male/aryan/hifigan/generator filter=lfs diff=lfs merge=lfs -text
+vocoder/male/dravidian/hifigan/generator filter=lfs diff=lfs merge=lfs -text

environment.yml

@@ -0,0 +1,127 @@
+name: tts-mfa-hifigan
+channels:
+  - defaults
+dependencies:
+  - _libgcc_mutex=0.1=main
+  - _openmp_mutex=5.1=1_gnu
+  - ca-certificates=2022.10.11=h06a4308_0
+  - certifi=2022.9.24=py37h06a4308_0
+  - ld_impl_linux-64=2.38=h1181459_1
+  - libffi=3.3=he6710b0_2
+  - libgcc-ng=11.2.0=h1234567_1
+  - libgomp=11.2.0=h1234567_1
+  - libstdcxx-ng=11.2.0=h1234567_1
+  - ncurses=6.3=h5eee18b_3
+  - openssl=1.1.1s=h7f8727e_0
+  - pip=22.2.2=py37h06a4308_0
+  - python=3.7.15=haa1d7c7_0
+  - readline=8.2=h5eee18b_0
+  - setuptools=65.5.0=py37h06a4308_0
+  - sqlite=3.39.3=h5082296_0
+  - tk=8.6.12=h1ccaba5_0
+  - wheel=0.37.1=pyhd3eb1b0_0
+  - xz=5.2.6=h5eee18b_0
+  - zlib=1.2.13=h5eee18b_0
+  - pip:
+    - aiosignal==1.3.1
+    - appdirs==1.4.4
+    - attrs==22.1.0
+    - audioread==3.0.0
+    - backcall==0.2.0
+    - cffi==1.15.1
+    - charset-normalizer==2.1.1
+    - ci-sdr==0.0.2
+    - click==8.0.4
+    - configargparse==1.5.3
+    - ctc-segmentation==1.7.4
+    - cycler==0.11.0
+    - cython==0.29.32
+    - decorator==5.1.1
+    - distance==0.1.3
+    - distlib==0.3.6
+    - docopt==0.6.2
+    - einops==0.6.0
+    - espnet==202209
+    - espnet-tts-frontend==0.0.3
+    - fast-bss-eval==0.1.3
+    - filelock==3.8.0
+    - flask==2.2.2
+    - fonttools==4.38.0
+    - frozenlist==1.3.3
+    - g2p-en==2.1.0
+    - grpcio==1.50.0
+    - gunicorn==20.1.0
+    - h5py==3.7.0
+    - humanfriendly==10.0
+    - idna==3.4
+    - importlib-metadata==4.13.0
+    - importlib-resources==5.10.0
+    - indic-num2words==1.0.1
+    - indic_unified_parser==1.0.6
+    - inflect==6.0.2
+    - ipython==7.34.0
+    - itsdangerous==2.1.2
+    - jaconv==0.3
+    - jamo==0.4.1
+    - jedi==0.18.2
+    - jinja2==3.1.2
+    - joblib==1.2.0
+    - jsonschema==4.17.0
+    - kaldiio==2.17.2
+    - kiwisolver==1.4.4
+    - librosa==0.9.2
+    - llvmlite==0.39.1
+    - markupsafe==2.1.1
+    - matplotlib==3.5.3
+    - matplotlib-inline==0.1.6
+    - msgpack==1.0.4
+    - nltk==3.7
+    - numba==0.56.4
+    - numpy==1.21.6
+    - packaging==21.3
+    - pandas==1.3.5
+    - parso==0.8.3
+    - pexpect==4.8.0
+    - pickleshare==0.7.5
+    - pillow==9.3.0
+    - pkgutil-resolve-name==1.3.10
+    - platformdirs==2.5.4
+    - pooch==1.6.0
+    - prompt-toolkit==3.0.36
+    - protobuf==3.20.1
+    - ptyprocess==0.7.0
+    - pycparser==2.21
+    - pydantic==1.10.2
+    - pydub==0.25.1
+    - pygments==2.14.0
+    - pyparsing==3.0.9
+    - pypinyin==0.44.0
+    - pyrsistent==0.19.2
+    - python-dateutil==2.8.2
+    - pytorch-wpe==0.0.1
+    - pytz==2022.6
+    - pyworld==0.3.2
+    - pyyaml==6.0
+    - ray==2.1.0
+    - regex==2022.10.31
+    - requests==2.28.1
+    - resampy==0.4.2
+    - scikit-learn==1.0.2
+    - scipy==1.7.3
+    - sentencepiece==0.1.97
+    - six==1.16.0
+    - soundfile==0.11.0
+    - threadpoolctl==3.1.0
+    - torch-complex==0.4.3
+    - tqdm==4.64.1
+    - traitlets==5.8.0
+    - typeguard==2.13.3
+    - typing-extensions==4.4.0
+    - unidecode==1.3.6
+    - urllib3==1.26.12
+    - virtualenv==20.16.7
+    - wcwidth==0.2.5
+    - webvtt-py==0.4.6
+    - werkzeug==2.2.2
+    - zipp==3.10.0
+prefix: /speech/Apps/Flask_app_env/conda_dir/envs/tts-mfa-hifigan

phone_dict/punjabi

@@ -0,0 +1,40 @@
+ਦੀ dI
+ਪਿਛਲੇ piClE
+ਮੁਤਾਬਕ mutAbak
+ਜਾਰੀ jArI
+ਮੌਸਮ mousam
+ਬਰਸਾਤ barsAt
+ਅਜੇ ajE
+ਪੈਣ pऐnx
+ਤੇ tE
+ਵਿੱਚ wiwc
+ਤੋਂ toq
+ਵੀ wI
+ਉੱਤਰ uwtar
+ਜਾ jA
+ਥਾਵਾਂ thAwAq
+ਦੋ do
+ਇਨ੍ਹਾਂ inhAq
+ਭਾਰੀ BArI
+ਬਿਹਾਰ bihAr
+ਹਰਿਆਣਾ hariAnxA
+ਦੌਰ dour
+ਹੈ hऐ
+ਦਾ dA
+ਰਾਜਾਂ rAjAq
+ਪ੍ਰਦੇਸ਼ pradEsank
+ਸਕ sak
+ਪਰ par
+ਕੀਤਾ kItA
+ਘੱਟ घawtx
+ਮੱਧ mawध
+ਵਿਭਾਗ wiBAg
+ਬੇਸ਼ੱਕ bEsankawk
+ਮੀਂਹ mIqh
+ਕੁਝ kuJ
+ਸੰਭਾਵਨਾ sarBAwnA
+ਪੰਜਾਬ parjAb
+ਨਹੀਂ nahIq
+ਦਿਨਾਂ dinAq
+ਇਨਕਾਰ inkAr
+ਹੋਈ hoI

punjabi/female/model/config.yaml

@@ -0,0 +1,281 @@
+config: conf/tuning/train_fastspeech2.yaml
+print_config: false
+log_level: INFO
+dry_run: false
+iterator_type: sequence
+output_dir: exp/tts_train_fastspeech2_raw_char_None
+ngpu: 1
+seed: 0
+num_workers: 1
+num_att_plot: 3
+dist_backend: nccl
+dist_init_method: env://
+dist_world_size: 8
+dist_rank: 0
+local_rank: 0
+dist_master_addr: localhost
+dist_master_port: 42343
+dist_launcher: null
+multiprocessing_distributed: true
+unused_parameters: false
+sharded_ddp: false
+cudnn_enabled: true
+cudnn_benchmark: false
+cudnn_deterministic: true
+collect_stats: false
+write_collected_feats: false
+max_epoch: 1000
+patience: null
+val_scheduler_criterion:
+- valid
+- loss
+early_stopping_criterion:
+- valid
+- loss
+- min
+best_model_criterion:
+-   - valid
+    - loss
+    - min
+-   - train
+    - loss
+    - min
+keep_nbest_models: 5
+grad_clip: 1.0
+grad_clip_type: 2.0
+grad_noise: false
+accum_grad: 8
+no_forward_run: false
+resume: true
+train_dtype: float32
+use_amp: false
+log_interval: null
+use_tensorboard: true
+use_wandb: false
+wandb_project: null
+wandb_id: null
+wandb_entity: null
+wandb_name: null
+wandb_model_log_interval: -1
+detect_anomaly: false
+pretrain_path: null
+init_param: []
+ignore_init_mismatch: false
+freeze_param: []
+num_iters_per_epoch: 800
+batch_size: 20
+valid_batch_size: null
+batch_bins: 3000000
+valid_batch_bins: null
+train_shape_file:
+- exp/tts_stats_raw_char_None/train/text_shape.char
+- exp/tts_stats_raw_char_None/train/speech_shape
+valid_shape_file:
+- exp/tts_stats_raw_char_None/valid/text_shape.char
+- exp/tts_stats_raw_char_None/valid/speech_shape
+batch_type: numel
+valid_batch_type: null
+fold_length:
+- 150
+- 204800
+sort_in_batch: descending
+sort_batch: descending
+multiple_iterator: false
+chunk_length: 500
+chunk_shift_ratio: 0.5
+num_cache_chunks: 1024
+train_data_path_and_name_and_type:
+-   - dump/raw/tr_no_dev/text
+    - text
+    - text
+-   - exp/tts_train_raw_char_None/decode_use_teacher_forcingtrue_train.loss.ave/tr_no_dev/durations
+    - durations
+    - text_int
+-   - dump/raw/tr_no_dev/wav.scp
+    - speech
+    - sound
+valid_data_path_and_name_and_type:
+-   - dump/raw/dev/text
+    - text
+    - text
+-   - exp/tts_train_raw_char_None/decode_use_teacher_forcingtrue_train.loss.ave/dev/durations
+    - durations
+    - text_int
+-   - dump/raw/dev/wav.scp
+    - speech
+    - sound
+allow_variable_data_keys: false
+max_cache_size: 0.0
+max_cache_fd: 32
+valid_max_cache_size: null
+optim: adam
+optim_conf:
+    lr: 1.0
+scheduler: noamlr
+scheduler_conf:
+    model_size: 384
+    warmup_steps: 4000
+token_list:
+- <blank>
+- <unk>
+- <space>
+- ਾ
+- ੇ
+- ੀ
+- ਰ
+- ਹ
+- ਿ
+- ਦ
+- ਂ
+- ਕ
+- ਸ
+- ਨ
+- ਲ
+- ਤ
+- ਆ
+- ਵ
+- ਮ
+- ੰ
+- ੱ
+- .
+- ੋ
+- ਪ
+- ਗ
+- ਜ
+- ੁ
+- ੂ
+- ਬ
+- ਚ
+- ਉ
+- ਣ
+- ੈ
+- ','
+- ਖ
+- ਇ
+- ਅ
+- ਈ
+- ੜ
+- ਟ
+- ਡ
+- ਸ਼
+- ੍
+- ਫ
+- ਭ
+- ਘ
+- ਏ
+- ਧ
+- ਥ
+- ਛ
+- ਜ਼
+- ਠ
+- ਝ
+- ੌ
+- '?'
+- ਐ
+- ਢ
+- ਼
+- ਓ
+- ਯ
+- '!'
+- ਲ਼
+- ਊ
+- ਗ਼
+- ਫ਼
+- ਔ
+- ਖ਼
+- ਃ
+- ੳ
+- ਞ
+- ੲ
+- ਙ
+- <sos/eos>
+odim: null
+model_conf: {}
+use_preprocessor: true
+token_type: char
+bpemodel: null
+non_linguistic_symbols: null
+cleaner: null
+g2p: g2p_en_no_space
+feats_extract: fbank
+feats_extract_conf:
+    n_fft: 1024
+    hop_length: 256
+    win_length: null
+    fs: 22050
+    fmin: 0
+    fmax: 8000
+    n_mels: 80
+normalize: global_mvn
+normalize_conf:
+    stats_file: /speech/arun/released_models/tts/female/punjabi/fastspeech2_tf_char/feats_stats.npz
+tts: fastspeech2
+tts_conf:
+    adim: 384
+    aheads: 2
+    elayers: 4
+    eunits: 1536
+    dlayers: 4
+    dunits: 1536
+    positionwise_layer_type: conv1d
+    positionwise_conv_kernel_size: 3
+    duration_predictor_layers: 2
+    duration_predictor_chans: 256
+    duration_predictor_kernel_size: 3
+    postnet_layers: 5
+    postnet_filts: 5
+    postnet_chans: 256
+    use_masking: true
+    use_scaled_pos_enc: true
+    encoder_normalize_before: true
+    decoder_normalize_before: true
+    reduction_factor: 1
+    init_type: xavier_uniform
+    init_enc_alpha: 1.0
+    init_dec_alpha: 1.0
+    transformer_enc_dropout_rate: 0.2
+    transformer_enc_positional_dropout_rate: 0.2
+    transformer_enc_attn_dropout_rate: 0.2
+    transformer_dec_dropout_rate: 0.2
+    transformer_dec_positional_dropout_rate: 0.2
+    transformer_dec_attn_dropout_rate: 0.2
+    pitch_predictor_layers: 5
+    pitch_predictor_chans: 256
+    pitch_predictor_kernel_size: 5
+    pitch_predictor_dropout: 0.5
+    pitch_embed_kernel_size: 1
+    pitch_embed_dropout: 0.0
+    stop_gradient_from_pitch_predictor: true
+    energy_predictor_layers: 2
+    energy_predictor_chans: 256
+    energy_predictor_kernel_size: 3
+    energy_predictor_dropout: 0.5
+    energy_embed_kernel_size: 1
+    energy_embed_dropout: 0.0
+    stop_gradient_from_energy_predictor: false
+pitch_extract: dio
+pitch_extract_conf:
+    fs: 22050
+    n_fft: 1024
+    hop_length: 256
+    f0max: 400
+    f0min: 80
+    reduction_factor: 1
+pitch_normalize: global_mvn
+pitch_normalize_conf:
+    stats_file: /speech/arun/released_models/tts/female/punjabi/fastspeech2_tf_char/pitch_stats.npz
+energy_extract: energy
+energy_extract_conf:
+    fs: 22050
+    n_fft: 1024
+    hop_length: 256
+    win_length: null
+    reduction_factor: 1
+energy_normalize: global_mvn
+energy_normalize_conf:
+    stats_file: /speech/arun/released_models/tts/female/punjabi/fastspeech2_tf_char/energy_stats.npz
+required:
+- output_dir
+- token_list
+version: 0.10.3a3
+distributed: true

punjabi/female/model/energy_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd2fe4ec1eb154c2796ef01f6fb615571ea4c1c516555dd0bf5160813e914d34
+size 770

punjabi/female/model/feats_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d0fa788f921881f4bf0f38779b1e80afd36397fa3091dc001a5efc89603f10b6
+size 1402

punjabi/female/model/feats_type

@@ -0,0 +1 @@
+raw

punjabi/female/model/model.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:557c40a10ac5f1a6da729b7888fa892c8248f3a25d54eb15afbca53b0a574203
+size 148718075

punjabi/female/model/pitch_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2cd742bfedeade855aa5b6fe9be805ba22e33111ae1204df4d6c7b8e662dcfee
+size 770

punjabi/male/model/config.yaml

@@ -0,0 +1,281 @@
+accum_grad: 8
+allow_variable_data_keys: false
+batch_bins: 3000000
+batch_size: 20
+batch_type: numel
+best_model_criterion:
+- - valid
+  - loss
+  - min
+- - train
+  - loss
+  - min
+bpemodel: null
+chunk_length: 500
+chunk_shift_ratio: 0.5
+cleaner: null
+collect_stats: false
+config: conf/tuning/train_fastspeech2.yaml
+cudnn_benchmark: false
+cudnn_deterministic: true
+cudnn_enabled: true
+detect_anomaly: false
+dist_backend: nccl
+dist_init_method: env://
+dist_launcher: null
+dist_master_addr: localhost
+dist_master_port: 37725
+dist_rank: 0
+dist_world_size: 8
+distributed: true
+dry_run: false
+early_stopping_criterion:
+- valid
+- loss
+- min
+energy_extract: energy
+energy_extract_conf:
+  fs: 22050
+  hop_length: 256
+  n_fft: 1024
+  reduction_factor: 1
+  win_length: null
+energy_normalize: global_mvn
+energy_normalize_conf:
+  stats_file: /home/speech/Fastspeech2_HS/punjabi/male/model/energy_stats.npz
+feats_extract: fbank
+feats_extract_conf:
+  fmax: 8000
+  fmin: 0
+  fs: 22050
+  hop_length: 256
+  n_fft: 1024
+  n_mels: 80
+  win_length: null
+fold_length:
+- 150
+- 204800
+freeze_param: []
+g2p: g2p_en_no_space
+grad_clip: 1.0
+grad_clip_type: 2.0
+grad_noise: false
+ignore_init_mismatch: false
+init_param: []
+iterator_type: sequence
+keep_nbest_models: 5
+local_rank: 0
+log_interval: null
+log_level: INFO
+max_cache_fd: 32
+max_cache_size: 0.0
+max_epoch: 1000
+model_conf: {}
+multiple_iterator: false
+multiprocessing_distributed: true
+ngpu: 1
+no_forward_run: false
+non_linguistic_symbols: null
+normalize: global_mvn
+normalize_conf:
+  stats_file: /home/speech/Fastspeech2_HS/punjabi/male/model/feats_stats.npz
+num_att_plot: 3
+num_cache_chunks: 1024
+num_iters_per_epoch: 800
+num_workers: 1
+odim: null
+optim: adam
+optim_conf:
+  lr: 1.0
+output_dir: exp/tts_train_fastspeech2_raw_char_None
+patience: null
+pitch_extract: dio
+pitch_extract_conf:
+  f0max: 400
+  f0min: 40
+  fs: 22050
+  hop_length: 256
+  n_fft: 1024
+  reduction_factor: 1
+pitch_normalize: global_mvn
+pitch_normalize_conf:
+  stats_file: /home/speech/Fastspeech2_HS/punjabi/male/model/pitch_stats.npz
+pretrain_path: null
+print_config: false
+required:
+- output_dir
+- token_list
+resume: true
+scheduler: noamlr
+scheduler_conf:
+  model_size: 384
+  warmup_steps: 4000
+seed: 0
+sharded_ddp: false
+sort_batch: descending
+sort_in_batch: descending
+token_list:
+- <blank>
+- <unk>
+- <space>
+- "\u0A3E"
+- "\u0A47"
+- "\u0A40"
+- "\u0A30"
+- "\u0A39"
+- "\u0A3F"
+- "\u0A26"
+- "\u0A02"
+- "\u0A15"
+- "\u0A38"
+- "\u0A28"
+- "\u0A32"
+- "\u0A24"
+- "\u0A06"
+- "\u0A35"
+- "\u0A2E"
+- "\u0A70"
+- "\u0A71"
+- .
+- "\u0A4B"
+- "\u0A2A"
+- "\u0A17"
+- "\u0A1C"
+- "\u0A41"
+- "\u0A42"
+- "\u0A2C"
+- "\u0A1A"
+- "\u0A09"
+- "\u0A23"
+- "\u0A48"
+- ','
+- "\u0A16"
+- "\u0A07"
+- "\u0A05"
+- "\u0A08"
+- "\u0A5C"
+- "\u0A1F"
+- "\u0A21"
+- "\u0A36"
+- "\u0A4D"
+- "\u0A2B"
+- "\u0A2D"
+- "\u0A18"
+- "\u0A0F"
+- "\u0A27"
+- "\u0A25"
+- "\u0A1B"
+- "\u0A5B"
+- "\u0A20"
+- "\u0A1D"
+- "\u0A4C"
+- '?'
+- "\u0A10"
+- "\u0A22"
+- "\u0A3C"
+- "\u0A13"
+- "\u0A2F"
+- '!'
+- "\u0A33"
+- "\u0A0A"
+- "\u0A5A"
+- "\u0A5E"
+- "\u0A14"
+- "\u0A59"
+- "\u0A03"
+- "\u0A73"
+- "\u0A1E"
+- "\u0A72"
+- "\u0A19"
+- <sos/eos>
+token_type: char
+train_data_path_and_name_and_type:
+- - dump/raw/tr_no_dev/text
+  - text
+  - text
+- - duration_info_from_teacher/decode_use_teacher_forcingtrue_train.loss.ave/tr_no_dev/durations
+  - durations
+  - text_int
+- - dump/raw/tr_no_dev/wav.scp
+  - speech
+  - sound
+train_dtype: float32
+train_shape_file:
+- exp/tts_stats_raw_char_None/train/text_shape.char
+- exp/tts_stats_raw_char_None/train/speech_shape
+tts: fastspeech2
+tts_conf:
+  adim: 384
+  aheads: 2
+  decoder_normalize_before: true
+  dlayers: 4
+  dunits: 1536
+  duration_predictor_chans: 256
+  duration_predictor_kernel_size: 3
+  duration_predictor_layers: 2
+  elayers: 4
+  encoder_normalize_before: true
+  energy_embed_dropout: 0.0
+  energy_embed_kernel_size: 1
+  energy_predictor_chans: 256
+  energy_predictor_dropout: 0.5
+  energy_predictor_kernel_size: 3
+  energy_predictor_layers: 2
+  eunits: 1536
+  init_dec_alpha: 1.0
+  init_enc_alpha: 1.0
+  init_type: xavier_uniform
+  pitch_embed_dropout: 0.0
+  pitch_embed_kernel_size: 1
+  pitch_predictor_chans: 256
+  pitch_predictor_dropout: 0.5
+  pitch_predictor_kernel_size: 5
+  pitch_predictor_layers: 5
+  positionwise_conv_kernel_size: 3
+  positionwise_layer_type: conv1d
+  postnet_chans: 256
+  postnet_filts: 5
+  postnet_layers: 5
+  reduction_factor: 1
+  stop_gradient_from_energy_predictor: false
+  stop_gradient_from_pitch_predictor: true
+  transformer_dec_attn_dropout_rate: 0.2
+  transformer_dec_dropout_rate: 0.2
+  transformer_dec_positional_dropout_rate: 0.2
+  transformer_enc_attn_dropout_rate: 0.2
+  transformer_enc_dropout_rate: 0.2
+  transformer_enc_positional_dropout_rate: 0.2
+  use_masking: true
+  use_scaled_pos_enc: true
+unused_parameters: false
+use_amp: false
+use_preprocessor: true
+use_tensorboard: true
+use_wandb: false
+val_scheduler_criterion:
+- valid
+- loss
+valid_batch_bins: null
+valid_batch_size: null
+valid_batch_type: null
+valid_data_path_and_name_and_type:
+- - dump/raw/dev/text
+  - text
+  - text
+- - duration_info_from_teacher/decode_use_teacher_forcingtrue_train.loss.ave/dev/durations
+  - durations
+  - text_int
+- - dump/raw/dev/wav.scp
+  - speech
+  - sound
+valid_max_cache_size: null
+valid_shape_file:
+- exp/tts_stats_raw_char_None/valid/text_shape.char
+- exp/tts_stats_raw_char_None/valid/speech_shape
+version: 0.10.3a3
+wandb_entity: null
+wandb_id: null
+wandb_model_log_interval: -1
+wandb_name: null
+wandb_project: null
+write_collected_feats: false

punjabi/male/model/energy_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:325fb34d8db2ddda8c4b4d3564227ee9fdc4338272e74c7943eb51567702a2ee
+size 770

punjabi/male/model/feats_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3cd21402e685536d58867c03dcb0a83baafc23d26ec3c1f20eaa0211a09351e2
+size 1402

punjabi/male/model/feats_type

@@ -0,0 +1 @@
+raw

punjabi/male/model/model.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:04260a1a75a05b97db3e8c23ecb745b96e902fec5760b67d1197a9caccf3e81b
+size 148718067

punjabi/male/model/pitch_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a75ab3091c7963b2f8fedeed769c0ddbd30c9106351d678a3a44dc4ad2eb9ffa
+size 770

rajasthani/female/model/config.yaml

@@ -0,0 +1,274 @@
+config: conf/tuning/train_fastspeech2.yaml
+print_config: false
+log_level: INFO
+dry_run: false
+iterator_type: sequence
+output_dir: exp/tts_train_fastspeech2_raw_char_None
+ngpu: 1
+seed: 0
+num_workers: 1
+num_att_plot: 3
+dist_backend: nccl
+dist_init_method: env://
+dist_world_size: 4
+dist_rank: 0
+local_rank: 0
+dist_master_addr: localhost
+dist_master_port: 60003
+dist_launcher: null
+multiprocessing_distributed: true
+unused_parameters: false
+sharded_ddp: false
+cudnn_enabled: true
+cudnn_benchmark: false
+cudnn_deterministic: true
+collect_stats: false
+write_collected_feats: false
+max_epoch: 1000
+patience: null
+val_scheduler_criterion:
+- valid
+- loss
+early_stopping_criterion:
+- valid
+- loss
+- min
+best_model_criterion:
+-   - valid
+    - loss
+    - min
+-   - train
+    - loss
+    - min
+keep_nbest_models: 5
+grad_clip: 1.0
+grad_clip_type: 2.0
+grad_noise: false
+accum_grad: 8
+no_forward_run: false
+resume: true
+train_dtype: float32
+use_amp: false
+log_interval: null
+use_tensorboard: true
+use_wandb: false
+wandb_project: null
+wandb_id: null
+wandb_entity: null
+wandb_name: null
+wandb_model_log_interval: -1
+detect_anomaly: false
+pretrain_path: null
+init_param: []
+ignore_init_mismatch: false
+freeze_param: []
+num_iters_per_epoch: 800
+batch_size: 20
+valid_batch_size: null
+batch_bins: 3000000
+valid_batch_bins: null
+train_shape_file:
+- exp/tts_stats_raw_char_None/train/text_shape.char
+- exp/tts_stats_raw_char_None/train/speech_shape
+valid_shape_file:
+- exp/tts_stats_raw_char_None/valid/text_shape.char
+- exp/tts_stats_raw_char_None/valid/speech_shape
+batch_type: numel
+valid_batch_type: null
+fold_length:
+- 150
+- 204800
+sort_in_batch: descending
+sort_batch: descending
+multiple_iterator: false
+chunk_length: 500
+chunk_shift_ratio: 0.5
+num_cache_chunks: 1024
+train_data_path_and_name_and_type:
+-   - dump/raw/tr_no_dev/text
+    - text
+    - text
+-   - duration_info/tr_no_dev/durations
+    - durations
+    - text_int
+-   - dump/raw/tr_no_dev/wav.scp
+    - speech
+    - sound
+-   - exp/tts_stats_raw_char_None/train/collect_feats/pitch.scp
+    - pitch
+    - npy
+-   - exp/tts_stats_raw_char_None/train/collect_feats/energy.scp
+    - energy
+    - npy
+valid_data_path_and_name_and_type:
+-   - dump/raw/dev/text
+    - text
+    - text
+-   - duration_info/dev/durations
+    - durations
+    - text_int
+-   - dump/raw/dev/wav.scp
+    - speech
+    - sound
+-   - exp/tts_stats_raw_char_None/valid/collect_feats/pitch.scp
+    - pitch
+    - npy
+-   - exp/tts_stats_raw_char_None/valid/collect_feats/energy.scp
+    - energy
+    - npy
+allow_variable_data_keys: false
+max_cache_size: 0.0
+max_cache_fd: 32
+valid_max_cache_size: null
+optim: adam
+optim_conf:
+    lr: 1.0
+scheduler: noamlr
+scheduler_conf:
+    model_size: 384
+    warmup_steps: 4000
+token_list:
+- <blank>
+- <unk>
+- A
+- a
+- r
+- ','
+- I
+- q
+- k
+- s
+- E
+- o
+- m
+- n
+- i
+- y
+- t
+- ऐ
+- w
+- j
+- p
+- h
+- l
+- d
+- b
+- ण
+- $
+- .
+- g
+- u
+- U
+- c
+- ख
+- ट
+- D
+- B
+- थ
+- ध
+- ള
+- श
+- औ
+- ठ
+- ड
+- C
+- ष
+- घ
+- P
+- J
+- ढ
+- z
+- f
+- R
+- ऑ
+- <sos/eos>
+odim: null
+model_conf: {}
+use_preprocessor: true
+token_type: char
+bpemodel: null
+non_linguistic_symbols: null
+cleaner: null
+g2p: g2p_en_no_space
+feats_extract: fbank
+feats_extract_conf:
+    n_fft: 1024
+    hop_length: 256
+    win_length: 1024
+    fs: 22050
+    fmin: 0
+    fmax: 8000
+    n_mels: 80
+normalize: global_mvn
+normalize_conf:
+    stats_file: /speech/arun/released_models/tts/female/rajasthani/fastspeech2_hs/feats_stats.npz
+tts: fastspeech2
+tts_conf:
+    adim: 384
+    aheads: 2
+    elayers: 4
+    eunits: 1536
+    dlayers: 4
+    dunits: 1536
+    positionwise_layer_type: conv1d
+    positionwise_conv_kernel_size: 3
+    duration_predictor_layers: 2
+    duration_predictor_chans: 256
+    duration_predictor_kernel_size: 3
+    postnet_layers: 5
+    postnet_filts: 5
+    postnet_chans: 256
+    use_masking: true
+    use_scaled_pos_enc: true
+    encoder_normalize_before: true
+    decoder_normalize_before: true
+    reduction_factor: 1
+    init_type: xavier_uniform
+    init_enc_alpha: 1.0
+    init_dec_alpha: 1.0
+    transformer_enc_dropout_rate: 0.2
+    transformer_enc_positional_dropout_rate: 0.2
+    transformer_enc_attn_dropout_rate: 0.2
+    transformer_dec_dropout_rate: 0.2
+    transformer_dec_positional_dropout_rate: 0.2
+    transformer_dec_attn_dropout_rate: 0.2
+    pitch_predictor_layers: 5
+    pitch_predictor_chans: 256
+    pitch_predictor_kernel_size: 5
+    pitch_predictor_dropout: 0.5
+    pitch_embed_kernel_size: 1
+    pitch_embed_dropout: 0.0
+    stop_gradient_from_pitch_predictor: true
+    energy_predictor_layers: 2
+    energy_predictor_chans: 256
+    energy_predictor_kernel_size: 3
+    energy_predictor_dropout: 0.5
+    energy_embed_kernel_size: 1
+    energy_embed_dropout: 0.0
+    stop_gradient_from_energy_predictor: false
+pitch_extract: dio
+pitch_extract_conf:
+    fs: 22050
+    n_fft: 1024
+    hop_length: 256
+    f0max: 400
+    f0min: 80
+    reduction_factor: 1
+pitch_normalize: global_mvn
+pitch_normalize_conf:
+    stats_file: /speech/arun/released_models/tts/female/rajasthani/fastspeech2_hs/pitch_stats.npz
+energy_extract: energy
+energy_extract_conf:
+    fs: 22050
+    n_fft: 1024
+    hop_length: 256
+    win_length: 1024
+    reduction_factor: 1
+energy_normalize: global_mvn
+energy_normalize_conf:
+    stats_file: /speech/arun/released_models/tts/female/rajasthani/fastspeech2_hs/energy_stats.npz
+required:
+- output_dir
+- token_list
+version: 0.10.3a3
+distributed: true

rajasthani/female/model/energy_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:92802cda100723b4dbd0a19a79f6df9f373d19425179c3d5304ad23af3247202
+size 770

rajasthani/female/model/feats_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:42ece0e29e4c2bca55edd5c3f23c22634312260c053bc73df657820e9f434790
+size 1402

rajasthani/female/model/feats_type

@@ -0,0 +1 @@
+raw

rajasthani/female/model/model.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ddce9f7517bb41a5fac33c7dcdd5a2de6413b8e12e4fb6278f27f12e3ff2cda4
+size 148688878

rajasthani/female/model/pitch_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:69bcb1b0a0d2f083545534fce050d081548ddc7d7479ca34c6d853d652be1c43
+size 770

rajasthani/male/model/config.yaml

@@ -0,0 +1,279 @@
+config: conf/tuning/train_fastspeech2.yaml
+print_config: false
+log_level: INFO
+dry_run: false
+iterator_type: sequence
+output_dir: exp/tts_train_fastspeech2_raw_char_None
+ngpu: 1
+seed: 0
+num_workers: 1
+num_att_plot: 3
+dist_backend: nccl
+dist_init_method: env://
+dist_world_size: 4
+dist_rank: 0
+local_rank: 0
+dist_master_addr: localhost
+dist_master_port: 43719
+dist_launcher: null
+multiprocessing_distributed: true
+unused_parameters: false
+sharded_ddp: false
+cudnn_enabled: true
+cudnn_benchmark: false
+cudnn_deterministic: true
+collect_stats: false
+write_collected_feats: false
+max_epoch: 1000
+patience: null
+val_scheduler_criterion:
+- valid
+- loss
+early_stopping_criterion:
+- valid
+- loss
+- min
+best_model_criterion:
+-   - valid
+    - loss
+    - min
+-   - train
+    - loss
+    - min
+keep_nbest_models: 5
+grad_clip: 1.0
+grad_clip_type: 2.0
+grad_noise: false
+accum_grad: 8
+no_forward_run: false
+resume: true
+train_dtype: float32
+use_amp: false
+log_interval: null
+use_tensorboard: true
+use_wandb: false
+wandb_project: null
+wandb_id: null
+wandb_entity: null
+wandb_name: null
+wandb_model_log_interval: -1
+detect_anomaly: false
+pretrain_path: null
+init_param: []
+ignore_init_mismatch: false
+freeze_param: []
+num_iters_per_epoch: 800
+batch_size: 20
+valid_batch_size: null
+batch_bins: 3000000
+valid_batch_bins: null
+train_shape_file:
+- exp/tts_stats_raw_char_None/train/text_shape.char
+- exp/tts_stats_raw_char_None/train/speech_shape
+valid_shape_file:
+- exp/tts_stats_raw_char_None/valid/text_shape.char
+- exp/tts_stats_raw_char_None/valid/speech_shape
+batch_type: numel
+valid_batch_type: null
+fold_length:
+- 150
+- 204800
+sort_in_batch: descending
+sort_batch: descending
+multiple_iterator: false
+chunk_length: 500
+chunk_shift_ratio: 0.5
+num_cache_chunks: 1024
+train_data_path_and_name_and_type:
+-   - dump/raw/tr_no_dev/text
+    - text
+    - text
+-   - duration_info/tr_no_dev/durations
+    - durations
+    - text_int
+-   - dump/raw/tr_no_dev/wav.scp
+    - speech
+    - sound
+-   - exp/tts_stats_raw_char_None/train/collect_feats/pitch.scp
+    - pitch
+    - npy
+-   - exp/tts_stats_raw_char_None/train/collect_feats/energy.scp
+    - energy
+    - npy
+valid_data_path_and_name_and_type:
+-   - dump/raw/dev/text
+    - text
+    - text
+-   - duration_info/dev/durations
+    - durations
+    - text_int
+-   - dump/raw/dev/wav.scp
+    - speech
+    - sound
+-   - exp/tts_stats_raw_char_None/valid/collect_feats/pitch.scp
+    - pitch
+    - npy
+-   - exp/tts_stats_raw_char_None/valid/collect_feats/energy.scp
+    - energy
+    - npy
+allow_variable_data_keys: false
+max_cache_size: 0.0
+max_cache_fd: 32
+valid_max_cache_size: null
+optim: adam
+optim_conf:
+    lr: 1.0
+scheduler: noamlr
+scheduler_conf:
+    model_size: 384
+    warmup_steps: 4000
+token_list:
+- <blank>
+- <unk>
+- A
+- a
+- r
+- I
+- ','
+- q
+- o
+- k
+- s
+- ऐ
+- m
+- E
+- i
+- n
+- y
+- t
+- w
+- j
+- h
+- l
+- p
+- d
+- ण
+- $
+- .
+- b
+- g
+- U
+- u
+- c
+- ख
+- D
+- ट
+- B
+- ध
+- थ
+- ള
+- ठ
+- औ
+- घ
+- श
+- ड
+- C
+- P
+- ष
+- J
+- T
+- ढ
+- z
+- R
+- ञ
+- f
+- ऑ
+- M
+- H
+- क
+- <sos/eos>
+odim: null
+model_conf: {}
+use_preprocessor: true
+token_type: char
+bpemodel: null
+non_linguistic_symbols: null
+cleaner: null
+g2p: g2p_en_no_space
+feats_extract: fbank
+feats_extract_conf:
+    n_fft: 1024
+    hop_length: 256
+    win_length: 1024
+    fs: 22050
+    fmin: 0
+    fmax: 8000
+    n_mels: 80
+normalize: global_mvn
+normalize_conf:
+    stats_file: /speech/arun/released_models/tts/male/rajasthani/fastspeech2_hs/feats_stats.npz
+tts: fastspeech2
+tts_conf:
+    adim: 384
+    aheads: 2
+    elayers: 4
+    eunits: 1536
+    dlayers: 4
+    dunits: 1536
+    positionwise_layer_type: conv1d
+    positionwise_conv_kernel_size: 3
+    duration_predictor_layers: 2
+    duration_predictor_chans: 256
+    duration_predictor_kernel_size: 3
+    postnet_layers: 5
+    postnet_filts: 5
+    postnet_chans: 256
+    use_masking: true
+    use_scaled_pos_enc: true
+    encoder_normalize_before: true
+    decoder_normalize_before: true
+    reduction_factor: 1
+    init_type: xavier_uniform
+    init_enc_alpha: 1.0
+    init_dec_alpha: 1.0
+    transformer_enc_dropout_rate: 0.2
+    transformer_enc_positional_dropout_rate: 0.2
+    transformer_enc_attn_dropout_rate: 0.2
+    transformer_dec_dropout_rate: 0.2
+    transformer_dec_positional_dropout_rate: 0.2
+    transformer_dec_attn_dropout_rate: 0.2
+    pitch_predictor_layers: 5
+    pitch_predictor_chans: 256
+    pitch_predictor_kernel_size: 5
+    pitch_predictor_dropout: 0.5
+    pitch_embed_kernel_size: 1
+    pitch_embed_dropout: 0.0
+    stop_gradient_from_pitch_predictor: true
+    energy_predictor_layers: 2
+    energy_predictor_chans: 256
+    energy_predictor_kernel_size: 3
+    energy_predictor_dropout: 0.5
+    energy_embed_kernel_size: 1
+    energy_embed_dropout: 0.0
+    stop_gradient_from_energy_predictor: false
+pitch_extract: dio
+pitch_extract_conf:
+    fs: 22050
+    n_fft: 1024
+    hop_length: 256
+    f0max: 350
+    f0min: 40
+    reduction_factor: 1
+pitch_normalize: global_mvn
+pitch_normalize_conf:
+    stats_file: /speech/arun/released_models/tts/male/rajasthani/fastspeech2_hs/pitch_stats.npz
+energy_extract: energy
+energy_extract_conf:
+    fs: 22050
+    n_fft: 1024
+    hop_length: 256
+    win_length: 1024
+    reduction_factor: 1
+energy_normalize: global_mvn
+energy_normalize_conf:
+    stats_file: /speech/arun/released_models/tts/male/rajasthani/fastspeech2_hs/energy_stats.npz
+required:
+- output_dir
+- token_list
+version: 0.10.3a3
+distributed: true

rajasthani/male/model/energy_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:de51706233506aafe9be85372f894ab3880065c9ece3d8006c203638567f12aa
+size 770

rajasthani/male/model/feats_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a23e2b255fed48edce9d6135410580eeeb4b965d6a76e814856cee6bed9ddace
+size 1402

rajasthani/male/model/feats_type

@@ -0,0 +1 @@
+raw

rajasthani/male/model/model.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:396e90e77e55643b1702345e1f2ffcaa95bc20243d94e64c1ffaef1484406678
+size 148696483

rajasthani/male/model/pitch_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1bb03e1a0f02a3f1a0f41fdffc01fdfad686ba393a5613f18961b1ac861b47b8
+size 770

tamil/female/model/config.yaml

@@ -0,0 +1,266 @@
+config: conf/tuning/train_fastspeech2.yaml
+print_config: false
+log_level: INFO
+dry_run: false
+iterator_type: sequence
+output_dir: exp/tts_train_fastspeech2_raw_char_None
+ngpu: 1
+seed: 0
+num_workers: 1
+num_att_plot: 3
+dist_backend: nccl
+dist_init_method: env://
+dist_world_size: 4
+dist_rank: 0
+local_rank: 0
+dist_master_addr: localhost
+dist_master_port: 46011
+dist_launcher: null
+multiprocessing_distributed: true
+unused_parameters: false
+sharded_ddp: false
+cudnn_enabled: true
+cudnn_benchmark: false
+cudnn_deterministic: true
+collect_stats: false
+write_collected_feats: false
+max_epoch: 1000
+patience: null
+val_scheduler_criterion:
+- valid
+- loss
+early_stopping_criterion:
+- valid
+- loss
+- min
+best_model_criterion:
+-   - valid
+    - loss
+    - min
+-   - train
+    - loss
+    - min
+keep_nbest_models: 5
+grad_clip: 1.0
+grad_clip_type: 2.0
+grad_noise: false
+accum_grad: 8
+no_forward_run: false
+resume: true
+train_dtype: float32
+use_amp: false
+log_interval: null
+use_tensorboard: true
+use_wandb: false
+wandb_project: null
+wandb_id: null
+wandb_entity: null
+wandb_name: null
+wandb_model_log_interval: -1
+detect_anomaly: false
+pretrain_path: null
+init_param: []
+ignore_init_mismatch: false
+freeze_param: []
+num_iters_per_epoch: 800
+batch_size: 20
+valid_batch_size: null
+batch_bins: 3000000
+valid_batch_bins: null
+train_shape_file:
+- exp/tts_stats_raw_char_None/train/text_shape.char
+- exp/tts_stats_raw_char_None/train/speech_shape
+valid_shape_file:
+- exp/tts_stats_raw_char_None/valid/text_shape.char
+- exp/tts_stats_raw_char_None/valid/speech_shape
+batch_type: numel
+valid_batch_type: null
+fold_length:
+- 150
+- 204800
+sort_in_batch: descending
+sort_batch: descending
+multiple_iterator: false
+chunk_length: 500
+chunk_shift_ratio: 0.5
+num_cache_chunks: 1024
+train_data_path_and_name_and_type:
+-   - dump/raw/tr_no_dev/text
+    - text
+    - text
+-   - duration_info/tr_no_dev/durations
+    - durations
+    - text_int
+-   - dump/raw/tr_no_dev/wav.scp
+    - speech
+    - sound
+-   - exp/tts_stats_raw_char_None/train/collect_feats/pitch.scp
+    - pitch
+    - npy
+-   - exp/tts_stats_raw_char_None/train/collect_feats/energy.scp
+    - energy
+    - npy
+valid_data_path_and_name_and_type:
+-   - dump/raw/dev/text
+    - text
+    - text
+-   - duration_info/dev/durations
+    - durations
+    - text_int
+-   - dump/raw/dev/wav.scp
+    - speech
+    - sound
+-   - exp/tts_stats_raw_char_None/valid/collect_feats/pitch.scp
+    - pitch
+    - npy
+-   - exp/tts_stats_raw_char_None/valid/collect_feats/energy.scp
+    - energy
+    - npy
+allow_variable_data_keys: false
+max_cache_size: 0.0
+max_cache_fd: 32
+valid_max_cache_size: null
+optim: adam
+optim_conf:
+    lr: 1.0
+scheduler: noamlr
+scheduler_conf:
+    model_size: 384
+    warmup_steps: 4000
+token_list:
+- <blank>
+- <unk>
+- a
+- i
+- ','
+- k
+- A
+- r
+- u
+- m
+- n
+- t
+- p
+- w
+- d
+- l
+- உ
+- y
+- g
+- ऐ
+- ट
+- न
+- ड
+- र
+- ള
+- s
+- e
+- E
+- ण
+- O
+- o
+- $
+- .
+- ङ
+- b
+- Z
+- U
+- I
+- c
+- j
+- ष
+- ञ
+- h
+- f
+- औ
+- <sos/eos>
+odim: null
+model_conf: {}
+use_preprocessor: true
+token_type: char
+bpemodel: null
+non_linguistic_symbols: null
+cleaner: null
+g2p: g2p_en_no_space
+feats_extract: fbank
+feats_extract_conf:
+    n_fft: 1024
+    hop_length: 256
+    win_length: 1024
+    fs: 22050
+    fmin: 0
+    fmax: 8000
+    n_mels: 80
+normalize: global_mvn
+normalize_conf:
+    stats_file: /speech/arun/released_models/tts/female/tamil/fastspeech2_hs/feats_stats.npz
+tts: fastspeech2
+tts_conf:
+    adim: 384
+    aheads: 2
+    elayers: 4
+    eunits: 1536
+    dlayers: 4
+    dunits: 1536
+    positionwise_layer_type: conv1d
+    positionwise_conv_kernel_size: 3
+    duration_predictor_layers: 2
+    duration_predictor_chans: 256
+    duration_predictor_kernel_size: 3
+    postnet_layers: 5
+    postnet_filts: 5
+    postnet_chans: 256
+    use_masking: true
+    use_scaled_pos_enc: true
+    encoder_normalize_before: true
+    decoder_normalize_before: true
+    reduction_factor: 1
+    init_type: xavier_uniform
+    init_enc_alpha: 1.0
+    init_dec_alpha: 1.0
+    transformer_enc_dropout_rate: 0.2
+    transformer_enc_positional_dropout_rate: 0.2
+    transformer_enc_attn_dropout_rate: 0.2
+    transformer_dec_dropout_rate: 0.2
+    transformer_dec_positional_dropout_rate: 0.2
+    transformer_dec_attn_dropout_rate: 0.2
+    pitch_predictor_layers: 5
+    pitch_predictor_chans: 256
+    pitch_predictor_kernel_size: 5
+    pitch_predictor_dropout: 0.5
+    pitch_embed_kernel_size: 1
+    pitch_embed_dropout: 0.0
+    stop_gradient_from_pitch_predictor: true
+    energy_predictor_layers: 2
+    energy_predictor_chans: 256
+    energy_predictor_kernel_size: 3
+    energy_predictor_dropout: 0.5
+    energy_embed_kernel_size: 1
+    energy_embed_dropout: 0.0
+    stop_gradient_from_energy_predictor: false
+pitch_extract: dio
+pitch_extract_conf:
+    fs: 22050
+    n_fft: 1024
+    hop_length: 256
+    f0max: 400
+    f0min: 80
+    reduction_factor: 1
+pitch_normalize: global_mvn
+pitch_normalize_conf:
+    stats_file: /speech/arun/released_models/tts/female/tamil/fastspeech2_hs/pitch_stats.npz
+energy_extract: energy
+energy_extract_conf:
+    fs: 22050
+    n_fft: 1024
+    hop_length: 256
+    win_length: 1024
+    reduction_factor: 1
+energy_normalize: global_mvn
+energy_normalize_conf:
+    stats_file: /speech/arun/released_models/tts/female/tamil/fastspeech2_hs/energy_stats.npz
+required:
+- output_dir
+- token_list
+version: 0.10.3a3
+distributed: true

tamil/female/model/energy_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd0aed3910609b6d14784a1ed884d2c4256283fc3102f3f0907f51d44905bc15
+size 770

tamil/female/model/feats_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:24a87e91f86da65483b0eee8e98abe8264019a133770a65a9fd3228d0ea7f277
+size 1402

tamil/female/model/feats_type

@@ -0,0 +1 @@
+raw

tamil/female/model/model.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:333232faa4ba548952eb9668f47bfb580cbe079643807d42113682ea887eade1
+size 148676597

tamil/female/model/pitch_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e01a98009365f40cad2d81a25dcbfd0a4194e99c700907da886cdca94229a214
+size 770

tamil/male/model/config.yaml

@@ -0,0 +1,272 @@
+accum_grad: 8
+allow_variable_data_keys: false
+batch_bins: 3000000
+batch_size: 20
+batch_type: numel
+best_model_criterion:
+- - valid
+  - loss
+  - min
+- - train
+  - loss
+  - min
+bpemodel: null
+chunk_length: 500
+chunk_shift_ratio: 0.5
+cleaner: null
+collect_stats: false
+config: conf/tuning/train_fastspeech2.yaml
+cudnn_benchmark: false
+cudnn_deterministic: true
+cudnn_enabled: true
+detect_anomaly: false
+dist_backend: nccl
+dist_init_method: env://
+dist_launcher: null
+dist_master_addr: localhost
+dist_master_port: 46753
+dist_rank: 0
+dist_world_size: 4
+distributed: true
+dry_run: false
+early_stopping_criterion:
+- valid
+- loss
+- min
+energy_extract: energy
+energy_extract_conf:
+  fs: 22050
+  hop_length: 256
+  n_fft: 1024
+  reduction_factor: 1
+  win_length: 1024
+energy_normalize: global_mvn
+energy_normalize_conf:
+  stats_file: /home/speech/Fastspeech2_HS/tamil/male/model/energy_stats.npz
+feats_extract: fbank
+feats_extract_conf:
+  fmax: 8000
+  fmin: 0
+  fs: 22050
+  hop_length: 256
+  n_fft: 1024
+  n_mels: 80
+  win_length: 1024
+fold_length:
+- 150
+- 204800
+freeze_param: []
+g2p: g2p_en_no_space
+grad_clip: 1.0
+grad_clip_type: 2.0
+grad_noise: false
+ignore_init_mismatch: false
+init_param: []
+iterator_type: sequence
+keep_nbest_models: 5
+local_rank: 0
+log_interval: null
+log_level: INFO
+max_cache_fd: 32
+max_cache_size: 0.0
+max_epoch: 1000
+model_conf: {}
+multiple_iterator: false
+multiprocessing_distributed: true
+ngpu: 1
+no_forward_run: false
+non_linguistic_symbols: null
+normalize: global_mvn
+normalize_conf:
+  stats_file: /home/speech/Fastspeech2_HS/tamil/male/model/feats_stats.npz
+num_att_plot: 3
+num_cache_chunks: 1024
+num_iters_per_epoch: 800
+num_workers: 1
+odim: null
+optim: adam
+optim_conf:
+  lr: 1.0
+output_dir: exp/tts_train_fastspeech2_raw_char_None
+patience: null
+pitch_extract: dio
+pitch_extract_conf:
+  f0max: 350
+  f0min: 40
+  fs: 22050
+  hop_length: 256
+  n_fft: 1024
+  reduction_factor: 1
+pitch_normalize: global_mvn
+pitch_normalize_conf:
+  stats_file: /home/speech/Fastspeech2_HS/tamil/male/model/pitch_stats.npz
+pretrain_path: null
+print_config: false
+required:
+- output_dir
+- token_list
+resume: true
+scheduler: noamlr
+scheduler_conf:
+  model_size: 384
+  warmup_steps: 4000
+seed: 0
+sharded_ddp: false
+sort_batch: descending
+sort_in_batch: descending
+token_list:
+- <blank>
+- <unk>
+- a
+- i
+- ','
+- r
+- n
+- "\u091F"
+- k
+- m
+- "\u0921"
+- d
+- w
+- l
+- A
+- s
+- E
+- p
+- u
+- t
+- "\u0910"
+- g
+- I
+- y
+- "\u0B89"
+- $
+- .
+- U
+- z
+- h
+- "\u0905"
+- f
+- b
+- o
+- "\u0928"
+- "\u0911"
+- "\u0930"
+- "\u0D33"
+- "\u0919"
+- e
+- "\u0923"
+- O
+- c
+- j
+- "\u0914"
+- "\u0936"
+- Z
+- "\u0937"
+- "\u091E"
+- C
+- "\u090D"
+- <sos/eos>
+token_type: char
+train_data_path_and_name_and_type:
+- - dump/raw/tr_no_dev/text
+  - text
+  - text
+- - duration_info/tr_no_dev/durations
+  - durations
+  - text_int
+- - dump/raw/tr_no_dev/wav.scp
+  - speech
+  - sound
+- - exp/tts_stats_raw_char_None/train/collect_feats/pitch.scp
+  - pitch
+  - npy
+- - exp/tts_stats_raw_char_None/train/collect_feats/energy.scp
+  - energy
+  - npy
+train_dtype: float32
+train_shape_file:
+- exp/tts_stats_raw_char_None/train/text_shape.char
+- exp/tts_stats_raw_char_None/train/speech_shape
+tts: fastspeech2
+tts_conf:
+  adim: 384
+  aheads: 2
+  decoder_normalize_before: true
+  dlayers: 4
+  dunits: 1536
+  duration_predictor_chans: 256
+  duration_predictor_kernel_size: 3
+  duration_predictor_layers: 2
+  elayers: 4
+  encoder_normalize_before: true
+  energy_embed_dropout: 0.0
+  energy_embed_kernel_size: 1
+  energy_predictor_chans: 256
+  energy_predictor_dropout: 0.5
+  energy_predictor_kernel_size: 3
+  energy_predictor_layers: 2
+  eunits: 1536
+  init_dec_alpha: 1.0
+  init_enc_alpha: 1.0
+  init_type: xavier_uniform
+  pitch_embed_dropout: 0.0
+  pitch_embed_kernel_size: 1
+  pitch_predictor_chans: 256
+  pitch_predictor_dropout: 0.5
+  pitch_predictor_kernel_size: 5
+  pitch_predictor_layers: 5
+  positionwise_conv_kernel_size: 3
+  positionwise_layer_type: conv1d
+  postnet_chans: 256
+  postnet_filts: 5
+  postnet_layers: 5
+  reduction_factor: 1
+  stop_gradient_from_energy_predictor: false
+  stop_gradient_from_pitch_predictor: true
+  transformer_dec_attn_dropout_rate: 0.2
+  transformer_dec_dropout_rate: 0.2
+  transformer_dec_positional_dropout_rate: 0.2
+  transformer_enc_attn_dropout_rate: 0.2
+  transformer_enc_dropout_rate: 0.2
+  transformer_enc_positional_dropout_rate: 0.2
+  use_masking: true
+  use_scaled_pos_enc: true
+unused_parameters: false
+use_amp: false
+use_preprocessor: true
+use_tensorboard: true
+use_wandb: false
+val_scheduler_criterion:
+- valid
+- loss
+valid_batch_bins: null
+valid_batch_size: null
+valid_batch_type: null
+valid_data_path_and_name_and_type:
+- - dump/raw/dev/text
+  - text
+  - text
+- - duration_info/dev/durations
+  - durations
+  - text_int
+- - dump/raw/dev/wav.scp
+  - speech
+  - sound
+- - exp/tts_stats_raw_char_None/valid/collect_feats/pitch.scp
+  - pitch
+  - npy
+- - exp/tts_stats_raw_char_None/valid/collect_feats/energy.scp
+  - energy
+  - npy
+valid_max_cache_size: null
+valid_shape_file:
+- exp/tts_stats_raw_char_None/valid/text_shape.char
+- exp/tts_stats_raw_char_None/valid/speech_shape
+version: 0.10.3a3
+wandb_entity: null
+wandb_id: null
+wandb_model_log_interval: -1
+wandb_name: null
+wandb_project: null
+write_collected_feats: false