first commit

.gitattributes

@@ -32,3 +32,7 @@ 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
+checkpoint_40000.pth filter=lfs diff=lfs merge=lfs -text
+supplemental/speakers-combined.pth filter=lfs diff=lfs merge=lfs -text
+supplemental/speakers-dataset.pth filter=lfs diff=lfs merge=lfs -text
+supplemental/model_se.pth.tar filter=lfs diff=lfs merge=lfs -text

checkpoint_40000.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:200b870bd53bf698babc890d0b47a4cf3c99246f58bc3825ffa6a9b4cb6353b1
+size 1043207501

config.json

@@ -0,0 +1,244 @@
+{
+    "logger_uri": null,
+    "project_name": "VCTK_Canadian_English",
+    "run_description": "",
+    "print_step": 50,
+    "plot_step": 100,
+    "model_param_stats": false,
+    "wandb_entity": null,
+    "dashboard_logger": "tensorboard",
+    "log_model_step": 1000,
+    "save_step": 5000,
+    "save_n_checkpoints": 2,
+    "save_checkpoints": true,
+    "save_all_best": false,
+    "save_best_after": 10000,
+    "target_loss": "loss_1",
+    "print_eval": false,
+    "test_delay_epochs": 0,
+    "run_eval": true,
+    "run_eval_steps": null,
+    "distributed_backend": "nccl",
+    "distributed_url": "tcp://localhost:54321",
+    "mixed_precision": true,
+    "epochs": 1000,
+    "batch_size": 16,
+    "eval_batch_size": 16,
+    "grad_clip": [
+        1000.0,
+        1000.0
+    ],
+    "scheduler_after_epoch": true,
+    "lr": 0.001,
+    "optimizer": "AdamW",
+    "optimizer_params": {
+        "betas": [
+            0.8,
+            0.99
+        ],
+        "eps": 1e-09,
+        "weight_decay": 0.01
+    },
+    "lr_scheduler": null,
+    "lr_scheduler_params": {},
+    "use_grad_scaler": false,
+    "cudnn_enable": true,
+    "cudnn_deterministic": false,
+    "cudnn_benchmark": false,
+    "training_seed": 54321,
+    "model": "vits",
+    "num_loader_workers": 8,
+    "num_eval_loader_workers": 0,
+    "use_noise_augment": false,
+    "audio": {
+        "fft_size": 1024,
+        "sample_rate": 16000,
+        "win_length": 1024,
+        "hop_length": 256,
+        "num_mels": 80,
+        "mel_fmin": 0,
+        "mel_fmax": null
+    },
+    "use_phonemes": false,
+    "phonemizer": "espeak",
+    "phoneme_language": "en-us",
+    "compute_input_seq_cache": true,
+    "text_cleaner": "multilingual_cleaners",
+    "enable_eos_bos_chars": false,
+    "test_sentences_file": "",
+    "characters": {
+        "characters_class": "TTS.tts.models.vits.VitsCharacters",
+        "vocab_dict": null,
+        "pad": "_",
+        "eos": "&",
+        "bos": "*",
+        "blank": null,
+        "characters": "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz\u00af\u00b7\u00df\u00e0\u00e1\u00e2\u00e3\u00e4\u00e6\u00e7\u00e8\u00e9\u00ea\u00eb\u00ec\u00ed\u00ee\u00ef\u00f1\u00f2\u00f3\u00f4\u00f5\u00f6\u00f9\u00fa\u00fb\u00fc\u00ff\u0101\u0105\u0107\u0113\u0119\u011b\u012b\u0131\u0142\u0144\u014d\u0151\u0153\u015b\u016b\u0171\u017a\u017c\u01ce\u01d0\u01d2\u01d4\u0430\u0431\u0432\u0433\u0434\u0435\u0436\u0437\u0438\u0439\u043a\u043b\u043c\u043d\u043e\u043f\u0440\u0441\u0442\u0443\u0444\u0445\u0446\u0447\u0448\u0449\u044a\u044b\u044c\u044d\u044e\u044f\u0451\u0454\u0456\u0457\u0491\u2013!'(),-.:;? ",
+        "punctuations": "!'(),-.:;? ",
+        "phonemes": "iy\u0268\u0289\u026fu\u026a\u028f\u028ae\u00f8\u0258\u0259\u0275\u0264o\u025b\u0153\u025c\u025e\u028c\u0254\u00e6\u0250a\u0276\u0251\u0252\u1d7b\u0298\u0253\u01c0\u0257\u01c3\u0284\u01c2\u0260\u01c1\u029bpbtd\u0288\u0256c\u025fk\u0261q\u0262\u0294\u0274\u014b\u0272\u0273n\u0271m\u0299r\u0280\u2c71\u027e\u027d\u0278\u03b2fv\u03b8\u00f0sz\u0283\u0292\u0282\u0290\u00e7\u029dx\u0263\u03c7\u0281\u0127\u0295h\u0266\u026c\u026e\u028b\u0279\u027bj\u0270l\u026d\u028e\u029f\u02c8\u02cc\u02d0\u02d1\u028dw\u0265\u029c\u02a2\u02a1\u0255\u0291\u027a\u0267\u025a\u02de\u026b'\u0303' ",
+        "is_unique": true,
+        "is_sorted": true
+    },
+    "add_blank": true,
+    "batch_group_size": 48,
+    "loss_masking": null,
+    "min_audio_len": 1,
+    "max_audio_len": 160000,
+    "min_text_len": 1,
+    "max_text_len": Infinity,
+    "compute_f0": false,
+    "compute_energy": false,
+    "compute_linear_spec": true,
+    "precompute_num_workers": 12,
+    "start_by_longest": true,
+    "shuffle": false,
+    "drop_last": false,
+    "datasets": [
+        {
+            "formatter": "vctk",
+            "language": "en"
+        }
+    ],
+    "eval_split_max_size": 256,
+    "eval_split_size": 0.01,
+    "use_speaker_weighted_sampler": false,
+    "speaker_weighted_sampler_alpha": 1.0,
+    "use_language_weighted_sampler": false,
+    "language_weighted_sampler_alpha": 1.0,
+    "use_length_weighted_sampler": false,
+    "length_weighted_sampler_alpha": 1.0,
+    "model_args": {
+        "num_chars": 284,
+        "out_channels": 513,
+        "spec_segment_size": 32,
+        "hidden_channels": 192,
+        "hidden_channels_ffn_text_encoder": 768,
+        "num_heads_text_encoder": 2,
+        "num_layers_text_encoder": 10,
+        "kernel_size_text_encoder": 3,
+        "dropout_p_text_encoder": 0.1,
+        "dropout_p_duration_predictor": 0.5,
+        "kernel_size_posterior_encoder": 5,
+        "dilation_rate_posterior_encoder": 1,
+        "num_layers_posterior_encoder": 16,
+        "kernel_size_flow": 5,
+        "dilation_rate_flow": 1,
+        "num_layers_flow": 4,
+        "resblock_type_decoder": "2",
+        "resblock_kernel_sizes_decoder": [
+            3,
+            7,
+            11
+        ],
+        "resblock_dilation_sizes_decoder": [
+            [
+                1,
+                3,
+                5
+            ],
+            [
+                1,
+                3,
+                5
+            ],
+            [
+                1,
+                3,
+                5
+            ]
+        ],
+        "upsample_rates_decoder": [
+            8,
+            8,
+            2,
+            2
+        ],
+        "upsample_initial_channel_decoder": 512,
+        "upsample_kernel_sizes_decoder": [
+            16,
+            16,
+            4,
+            4
+        ],
+        "periods_multi_period_discriminator": [
+            2,
+            3,
+            5,
+            7,
+            11
+        ],
+        "use_sdp": true,
+        "noise_scale": 1.0,
+        "inference_noise_scale": 0.667,
+        "length_scale": 1.0,
+        "noise_scale_dp": 1.0,
+        "inference_noise_scale_dp": 1.0,
+        "max_inference_len": null,
+        "init_discriminator": true,
+        "use_spectral_norm_disriminator": false,
+        "use_speaker_embedding": false,
+        "num_speakers": 14,
+        "speakers_file": "supplemental/speakers.pth",
+        "d_vector_file": [
+            "supplemental/speakers-base.json",
+            "supplemental/speakers-dataset.pth"
+        ],
+        "speaker_embedding_channels": 256,
+        "use_d_vector_file": true,
+        "d_vector_dim": 512,
+        "detach_dp_input": true,
+        "use_language_embedding": true,
+        "embedded_language_dim": 4,
+        "num_languages": 3,
+        "language_ids_file": "supplemental/language_ids.json",
+        "use_speaker_encoder_as_loss": true,
+        "speaker_encoder_config_path": "supplemental/config_se.json",
+        "speaker_encoder_model_path": "supplemental/model_se.pth.tar",
+        "condition_dp_on_speaker": true,
+        "freeze_encoder": false,
+        "freeze_DP": false,
+        "freeze_PE": false,
+        "freeze_flow_decoder": false,
+        "freeze_waveform_decoder": false,
+        "encoder_sample_rate": null,
+        "interpolate_z": true,
+        "reinit_DP": false,
+        "reinit_text_encoder": false
+    },
+    "lr_gen": 0.0002,
+    "lr_disc": 0.0002,
+    "lr_scheduler_gen": "ExponentialLR",
+    "lr_scheduler_gen_params": {
+        "gamma": 0.999875,
+        "last_epoch": -1
+    },
+    "lr_scheduler_disc": "ExponentialLR",
+    "lr_scheduler_disc_params": {
+        "gamma": 0.999875,
+        "last_epoch": -1
+    },
+    "kl_loss_alpha": 1.0,
+    "disc_loss_alpha": 1.0,
+    "gen_loss_alpha": 1.0,
+    "feat_loss_alpha": 1.0,
+    "mel_loss_alpha": 45.0,
+    "dur_loss_alpha": 1.0,
+    "speaker_encoder_loss_alpha": 9.0,
+    "return_wav": true,
+    "use_weighted_sampler": false,
+    "weighted_sampler_attrs": {},
+    "weighted_sampler_multipliers": {},
+    "r": 1,
+    "num_speakers": 14,
+    "use_speaker_embedding": false,
+    "speakers_file": "supplemental/speakers.pth",
+    "speaker_embedding_channels": 256,
+    "language_ids_file": "supplemental/language_ids.json",
+    "use_language_embedding": true,
+    "use_d_vector_file": true,
+    "d_vector_file": [
+        "supplemental/speakers-base.json",
+        "supplemental/speakers-dataset.pth"
+    ],
+    "d_vector_dim": 512
+}

prepare_model.py

@@ -0,0 +1,58 @@
+import json
+import os
+import subprocess
+
+# Load the data from the provided dictionary
+data = {
+        "VCTK_p303": {"age": 24, "gender": "F", "accents": "Canadian", "region": "Toronto", "comments": ""},
+        "VCTK_p312": {"age": 19, "gender": "F", "accents": "Canadian", "region": "Hamilton", "comments": ""},
+        "VCTK_p317": {"age": 23, "gender": "F", "accents": "Canadian", "region": "Hamilton", "comments": ""},
+        "VCTK_p343": {"age": 27, "gender": "F", "accents": "Canadian", "region": "Alberta", "comments": ""},
+        "VCTK_p307": {"age": 23, "gender": "F", "accents": "Canadian", "region": "Ontario", "comments": ""},
+        "VCTK_p316": {"age": 20, "gender": "M", "accents": "Canadian", "region": "Alberta", "comments": ""},
+        "VCTK_p363": {"age": 22, "gender": "M", "accents": "Canadian", "region": "Toronto", "comments": ""},
+        "VCTK_p302": {"age": 20, "gender": "M", "accents": "Canadian", "region": "Montreal", "comments": ""}
+    }
+
+
+# Convert the data to JSON format
+json_data = json.dumps(data, indent=2)
+
+# Save the JSON data to a file
+with open('speakers-log.json', 'w') as file:
+    file.write(json_data)
+
+# Run the TTS command to get the speaker indices
+command = "tts --model_path checkpoint_40000.pth --config_path config.json --list_speaker_idxs | grep -vE '^(\s*\||\s*>|\s*$)'"
+output = subprocess.check_output(command, shell=True, text=True)
+
+# Parse the JSON output into a Python dictionary
+speaker_indices = eval(output)
+
+# Load the speaker IDs from speakers.json
+with open('speakers-log.json', 'r') as file:
+    speaker_ids = json.load(file)
+
+
+for speaker_idx in speaker_indices:
+    # # Remove the 'VCTK_' prefix
+    speaker_id = speaker_idx
+    # speaker_id = speaker_idx.replace('VCTK_', '')
+
+    # Lookup the speaker ID in the loaded speaker IDs
+    if speaker_id in speaker_ids:
+        speaker_id_json = speaker_ids[speaker_id]
+    else:
+        continue
+
+    # # Generate the TTS command to create the audio file
+    text = f"Hello, I am from {speaker_id_json['region']}. I hope that you will select my voice for your project. Thank you."
+    # # make samples directory if it doesn't exist
+    if not os.path.exists("samples"):
+        os.makedirs("samples")
+
+    out_path = f"samples/{speaker_id}.wav"
+    tts_command = f"tts --text \"{text}\" --model_path checkpoint_40000.pth --language_idx en --config_path config.json --speaker_idx \"{speaker_id}\" --out_path {out_path}"
+
+    # Execute the TTS command
+    os.system(tts_command)

speakers-log.json

@@ -0,0 +1,58 @@
+{
+  "VCTK_p303": {
+    "age": 24,
+    "gender": "F",
+    "accents": "Canadian",
+    "region": "Toronto",
+    "comments": ""
+  },
+  "VCTK_p312": {
+    "age": 19,
+    "gender": "F",
+    "accents": "Canadian",
+    "region": "Hamilton",
+    "comments": ""
+  },
+  "VCTK_p317": {
+    "age": 23,
+    "gender": "F",
+    "accents": "Canadian",
+    "region": "Hamilton",
+    "comments": ""
+  },
+  "VCTK_p343": {
+    "age": 27,
+    "gender": "F",
+    "accents": "Canadian",
+    "region": "Alberta",
+    "comments": ""
+  },
+  "VCTK_p307": {
+    "age": 23,
+    "gender": "F",
+    "accents": "Canadian",
+    "region": "Ontario",
+    "comments": ""
+  },
+  "VCTK_p316": {
+    "age": 20,
+    "gender": "M",
+    "accents": "Canadian",
+    "region": "Alberta",
+    "comments": ""
+  },
+  "VCTK_p363": {
+    "age": 22,
+    "gender": "M",
+    "accents": "Canadian",
+    "region": "Toronto",
+    "comments": ""
+  },
+  "VCTK_p302": {
+    "age": 20,
+    "gender": "M",
+    "accents": "Canadian",
+    "region": "Montreal",
+    "comments": ""
+  }
+}

supplemental/config_se.json

@@ -0,0 +1,121 @@
+{
+    "model": "speaker_encoder",
+    "run_name": "speaker_encoder",
+    "run_description": "resnet speaker encoder trained with commonvoice all languages dev and train, Voxceleb 1 dev and Voxceleb 2 dev",
+    "epochs": 100000,
+    "batch_size": null,
+    "eval_batch_size": null,
+    "mixed_precision": false,
+    "run_eval": true,
+    "test_delay_epochs": 0,
+    "print_eval": false,
+    "print_step": 50,
+    "tb_plot_step": 100,
+    "tb_model_param_stats": false,
+    "save_step": 1000,
+    "checkpoint": true,
+    "keep_all_best": false,
+    "keep_after": 10000,
+    "num_loader_workers": 8,
+    "num_val_loader_workers": 0,
+    "use_noise_augment": false,
+    "output_path": "../checkpoints/speaker_encoder/language_balanced/normalized/angleproto-4-samples-by-speakers/",
+    "distributed_backend": "nccl",
+    "distributed_url": "tcp://localhost:54321",
+    "audio": {
+        "fft_size": 512,
+        "win_length": 400,
+        "hop_length": 160,
+        "frame_shift_ms": null,
+        "frame_length_ms": null,
+        "stft_pad_mode": "reflect",
+        "sample_rate": 16000,
+        "resample": false,
+        "preemphasis": 0.97,
+        "ref_level_db": 20,
+        "do_sound_norm": false,
+        "do_trim_silence": false,
+        "trim_db": 60,
+        "power": 1.5,
+        "griffin_lim_iters": 60,
+        "num_mels": 64,
+        "mel_fmin": 0.0,
+        "mel_fmax": 8000.0,
+        "spec_gain": 20,
+        "signal_norm": false,
+        "min_level_db": -100,
+        "symmetric_norm": false,
+        "max_norm": 4.0,
+        "clip_norm": false,
+        "stats_path": null,
+        "do_rms_norm": true,
+        "db_level": -27.0
+    },
+    "datasets": [
+        {
+            "name": "voxceleb2",
+            "path": "/workspace/scratch/ecasanova/datasets/VoxCeleb/vox2_dev_aac/",
+            "meta_file_train": null,
+            "ununsed_speakers": null,
+            "meta_file_val": null,
+            "meta_file_attn_mask": "",
+            "language": "voxceleb"
+        }
+    ],
+    "model_params": {
+        "model_name": "resnet",
+        "input_dim": 64,
+        "use_torch_spec": true,
+        "log_input": true,
+        "proj_dim": 512
+    },
+    "audio_augmentation": {
+        "p": 0.5,
+        "rir": {
+            "rir_path": "/workspace/store/ecasanova/ComParE/RIRS_NOISES/simulated_rirs/",
+            "conv_mode": "full"
+        },
+        "additive": {
+            "sounds_path": "/workspace/store/ecasanova/ComParE/musan/",
+            "speech": {
+                "min_snr_in_db": 13,
+                "max_snr_in_db": 20,
+                "min_num_noises": 1,
+                "max_num_noises": 1
+            },
+            "noise": {
+                "min_snr_in_db": 0,
+                "max_snr_in_db": 15,
+                "min_num_noises": 1,
+                "max_num_noises": 1
+            },
+            "music": {
+                "min_snr_in_db": 5,
+                "max_snr_in_db": 15,
+                "min_num_noises": 1,
+                "max_num_noises": 1
+            }
+        },
+        "gaussian": {
+            "p": 0.0,
+            "min_amplitude": 0.0,
+            "max_amplitude": 1e-05
+        }
+    },
+    "storage": {
+        "sample_from_storage_p": 0.5,
+        "storage_size": 40
+    },
+    "max_train_step": 1000000,
+    "loss": "angleproto",
+    "grad_clip": 3.0,
+    "lr": 0.0001,
+    "lr_decay": false,
+    "warmup_steps": 4000,
+    "wd": 1e-06,
+    "steps_plot_stats": 100,
+    "num_speakers_in_batch": 100,
+    "num_utters_per_speaker": 4,
+    "skip_speakers": true,
+    "voice_len": 2.0
+}

supplemental/language_ids.json

@@ -0,0 +1,5 @@
+{
+    "en": 0,
+    "fr-fr": 1,
+    "pt-br": 2
+}

supplemental/model_se.pth.tar

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8f96efb20cbeeefd81fd8336d7f0155bf8902f82f9474e58ccb19d9e12345172
+size 44610930

supplemental/speaker_ids.json

@@ -0,0 +1,16 @@
+{
+    "female-en-5": 0,
+    "female-en-5\n": 1,
+    "female-pt-4\n": 2,
+    "male-en-2": 3,
+    "male-en-2\n": 4,
+    "male-pt-3\n": 5,
+    "VCTK_p302": 6,
+    "VCTK_p303": 7,
+    "VCTK_p307": 8,
+    "VCTK_p312": 9,
+    "VCTK_p316": 10,
+    "VCTK_p317": 11,
+    "VCTK_p343": 12,
+    "VCTK_p363": 13
+}

supplemental/speakers-combined.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:becb5d22b96aeb6b93afa098f8e07af3ddefdd13279ee43ee6353171d4c9f303
+size 14983919

supplemental/speakers-dataset.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bb0ab1326f8373e9e9457f3be10b0181317ab6bf4ee15c9e79f46af737ab11fe
+size 14803055