Upload 16 files

fastspeech/config.yml

@@ -0,0 +1,79 @@
+# This is the hyperparameter configuration file for FastSpeech v3.
+# Please make sure this is adjusted for the LJSpeech dataset. If you want to
+# apply to the other dataset, you might need to carefully change some parameters.
+# This configuration performs 200k iters but a best checkpoint is around 150k iters.
+
+###########################################################
+#                FEATURE EXTRACTION SETTING               #
+###########################################################
+hop_size: 256            # Hop size.
+format: "npy"
+
+
+###########################################################
+#              NETWORK ARCHITECTURE SETTING               #
+###########################################################
+model_type: "fastspeech"
+
+fastspeech_params:
+    n_speakers: 1
+    encoder_hidden_size: 384
+    encoder_num_hidden_layers: 4
+    encoder_num_attention_heads: 2
+    encoder_attention_head_size: 192  # hidden_size // num_attention_heads
+    encoder_intermediate_size: 1024
+    encoder_intermediate_kernel_size: 3
+    encoder_hidden_act: "mish"
+    decoder_hidden_size: 384
+    decoder_num_hidden_layers: 4
+    decoder_num_attention_heads: 2
+    decoder_attention_head_size: 192  # hidden_size // num_attention_heads
+    decoder_intermediate_size: 1024
+    decoder_intermediate_kernel_size: 3
+    decoder_hidden_act: "mish"
+    num_duration_conv_layers: 2
+    duration_predictor_filters: 256
+    duration_predictor_kernel_sizes: 3
+    num_mels: 80
+    hidden_dropout_prob: 0.2
+    attention_probs_dropout_prob: 0.1
+    duration_predictor_dropout_probs: 0.2
+    max_position_embeddings: 2048
+    initializer_range: 0.02
+    output_attentions: False
+    output_hidden_states: False
+
+###########################################################
+#                  DATA LOADER SETTING                    #
+###########################################################
+batch_size: 16              # Batch size for each GPU with assuming that gradient_accumulation_steps == 1.
+remove_short_samples: true  # Whether to remove samples the length of which are less than batch_max_steps.
+allow_cache: true           # Whether to allow cache in dataset. If true, it requires cpu memory.
+mel_length_threshold: 32    # remove all targets has mel_length <= 32 
+is_shuffle: true            # shuffle dataset after each epoch.
+###########################################################
+#             OPTIMIZER & SCHEDULER SETTING               #
+###########################################################
+optimizer_params:
+    initial_learning_rate: 0.001
+    end_learning_rate: 0.00005
+    decay_steps: 150000          # < train_max_steps is recommend.
+    warmup_proportion: 0.02
+    weight_decay: 0.001
+
+gradient_accumulation_steps: 1
+var_train_expr: null  # trainable variable expr (eg. 'embeddings|encoder|decoder' )
+                      # must separate by |. if var_train_expr is null then we 
+                      # training all variable     
+###########################################################
+#                    INTERVAL SETTING                     #
+###########################################################
+train_max_steps: 200000               # Number of training steps.
+save_interval_steps: 5000             # Interval steps to save checkpoint.
+eval_interval_steps: 500              # Interval steps to evaluate the network.
+log_interval_steps: 200               # Interval steps to record the training log.
+
+###########################################################
+#                     OTHER SETTING                       #
+###########################################################
+num_save_intermediate_results: 1  # Number of batch to be saved as intermediate results.

fastspeech/model.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4d0352241536eec6bb426dfa8d69b8a5a96bb6ef51fe3a5599b68ec190b83a85
+size 120784120

fastspeech/processor.json

@@ -0,0 +1 @@
+{"symbol_to_id": {"pad": 0, "-": 1, "!": 2, "'": 3, "(": 4, ")": 5, ",": 6, ".": 7, ":": 8, ";": 9, "?": 10, " ": 11, "A": 12, "B": 13, "C": 14, "D": 15, "E": 16, "F": 17, "G": 18, "H": 19, "I": 20, "J": 21, "K": 22, "L": 23, "M": 24, "N": 25, "O": 26, "P": 27, "Q": 28, "R": 29, "S": 30, "T": 31, "U": 32, "V": 33, "W": 34, "X": 35, "Y": 36, "Z": 37, "a": 38, "b": 39, "c": 40, "d": 41, "e": 42, "f": 43, "g": 44, "h": 45, "i": 46, "j": 47, "k": 48, "l": 49, "m": 50, "n": 51, "o": 52, "p": 53, "q": 54, "r": 55, "s": 56, "t": 57, "u": 58, "v": 59, "w": 60, "x": 61, "y": 62, "z": 63, "@AA": 64, "@AA0": 65, "@AA1": 66, "@AA2": 67, "@AE": 68, "@AE0": 69, "@AE1": 70, "@AE2": 71, "@AH": 72, "@AH0": 73, "@AH1": 74, "@AH2": 75, "@AO": 76, "@AO0": 77, "@AO1": 78, "@AO2": 79, "@AW": 80, "@AW0": 81, "@AW1": 82, "@AW2": 83, "@AY": 84, "@AY0": 85, "@AY1": 86, "@AY2": 87, "@B": 88, "@CH": 89, "@D": 90, "@DH": 91, "@EH": 92, "@EH0": 93, "@EH1": 94, "@EH2": 95, "@ER": 96, "@ER0": 97, "@ER1": 98, "@ER2": 99, "@EY": 100, "@EY0": 101, "@EY1": 102, "@EY2": 103, "@F": 104, "@G": 105, "@HH": 106, "@IH": 107, "@IH0": 108, "@IH1": 109, "@IH2": 110, "@IY": 111, "@IY0": 112, "@IY1": 113, "@IY2": 114, "@JH": 115, "@K": 116, "@L": 117, "@M": 118, "@N": 119, "@NG": 120, "@OW": 121, "@OW0": 122, "@OW1": 123, "@OW2": 124, "@OY": 125, "@OY0": 126, "@OY1": 127, "@OY2": 128, "@P": 129, "@R": 130, "@S": 131, "@SH": 132, "@T": 133, "@TH": 134, "@UH": 135, "@UH0": 136, "@UH1": 137, "@UH2": 138, "@UW": 139, "@UW0": 140, "@UW1": 141, "@UW2": 142, "@V": 143, "@W": 144, "@Y": 145, "@Z": 146, "@ZH": 147, "eos": 148}, "id_to_symbol": {"0": "pad", "1": "-", "2": "!", "3": "'", "4": "(", "5": ")", "6": ",", "7": ".", "8": ":", "9": ";", "10": "?", "11": " ", "12": "A", "13": "B", "14": "C", "15": "D", "16": "E", "17": "F", "18": "G", "19": "H", "20": "I", "21": "J", "22": "K", "23": "L", "24": "M", "25": "N", "26": "O", "27": "P", "28": "Q", "29": "R", "30": "S", "31": "T", "32": "U", "33": "V", "34": "W", "35": "X", "36": "Y", "37": "Z", "38": "a", "39": "b", "40": "c", "41": "d", "42": "e", "43": "f", "44": "g", "45": "h", "46": "i", "47": "j", "48": "k", "49": "l", "50": "m", "51": "n", "52": "o", "53": "p", "54": "q", "55": "r", "56": "s", "57": "t", "58": "u", "59": "v", "60": "w", "61": "x", "62": "y", "63": "z", "64": "@AA", "65": "@AA0", "66": "@AA1", "67": "@AA2", "68": "@AE", "69": "@AE0", "70": "@AE1", "71": "@AE2", "72": "@AH", "73": "@AH0", "74": "@AH1", "75": "@AH2", "76": "@AO", "77": "@AO0", "78": "@AO1", "79": "@AO2", "80": "@AW", "81": "@AW0", "82": "@AW1", "83": "@AW2", "84": "@AY", "85": "@AY0", "86": "@AY1", "87": "@AY2", "88": "@B", "89": "@CH", "90": "@D", "91": "@DH", "92": "@EH", "93": "@EH0", "94": "@EH1", "95": "@EH2", "96": "@ER", "97": "@ER0", "98": "@ER1", "99": "@ER2", "100": "@EY", "101": "@EY0", "102": "@EY1", "103": "@EY2", "104": "@F", "105": "@G", "106": "@HH", "107": "@IH", "108": "@IH0", "109": "@IH1", "110": "@IH2", "111": "@IY", "112": "@IY0", "113": "@IY1", "114": "@IY2", "115": "@JH", "116": "@K", "117": "@L", "118": "@M", "119": "@N", "120": "@NG", "121": "@OW", "122": "@OW0", "123": "@OW1", "124": "@OW2", "125": "@OY", "126": "@OY0", "127": "@OY1", "128": "@OY2", "129": "@P", "130": "@R", "131": "@S", "132": "@SH", "133": "@T", "134": "@TH", "135": "@UH", "136": "@UH0", "137": "@UH1", "138": "@UH2", "139": "@UW", "140": "@UW0", "141": "@UW1", "142": "@UW2", "143": "@V", "144": "@W", "145": "@Y", "146": "@Z", "147": "@ZH", "148": "eos"}, "speakers_map": {"ljspeech": 0}, "processor_name": "LJSpeechProcessor"}

fastspeech2/config.yml

@@ -0,0 +1,78 @@
+# This is the hyperparameter configuration file for FastSpeech2 v1.
+# Please make sure this is adjusted for the LJSpeech dataset. If you want to
+# apply to the other dataset, you might need to carefully change some parameters.
+# This configuration performs 200k iters but a best checkpoint is around 150k iters.
+
+###########################################################
+#                FEATURE EXTRACTION SETTING               #
+###########################################################
+hop_size: 256            # Hop size.
+format: "npy"
+
+
+###########################################################
+#              NETWORK ARCHITECTURE SETTING               #
+###########################################################
+model_type: "fastspeech2"
+
+fastspeech2_params:
+    n_speakers: 1
+    encoder_hidden_size: 384
+    encoder_num_hidden_layers: 4
+    encoder_num_attention_heads: 2
+    encoder_attention_head_size: 192  # hidden_size // num_attention_heads
+    encoder_intermediate_size: 1024
+    encoder_intermediate_kernel_size: 3
+    encoder_hidden_act: "mish"
+    decoder_hidden_size: 384
+    decoder_num_hidden_layers: 4
+    decoder_num_attention_heads: 2
+    decoder_attention_head_size: 192  # hidden_size // num_attention_heads
+    decoder_intermediate_size: 1024
+    decoder_intermediate_kernel_size: 3
+    decoder_hidden_act: "mish"
+    variant_prediction_num_conv_layers: 2
+    variant_predictor_filter: 256
+    variant_predictor_kernel_size: 3
+    variant_predictor_dropout_rate: 0.5
+    num_mels: 80
+    hidden_dropout_prob: 0.2
+    attention_probs_dropout_prob: 0.1
+    max_position_embeddings: 2048
+    initializer_range: 0.02
+    output_attentions: False
+    output_hidden_states: False
+
+###########################################################
+#                  DATA LOADER SETTING                    #
+###########################################################
+batch_size: 16              # Batch size for each GPU with assuming that gradient_accumulation_steps == 1.
+remove_short_samples: true  # Whether to remove samples the length of which are less than batch_max_steps.
+allow_cache: true           # Whether to allow cache in dataset. If true, it requires cpu memory.
+mel_length_threshold: 32    # remove all targets has mel_length <= 32 
+is_shuffle: true            # shuffle dataset after each epoch.
+###########################################################
+#             OPTIMIZER & SCHEDULER SETTING               #
+###########################################################
+optimizer_params:
+    initial_learning_rate: 0.001
+    end_learning_rate: 0.00005
+    decay_steps: 150000          # < train_max_steps is recommend.
+    warmup_proportion: 0.02
+    weight_decay: 0.001
+
+gradient_accumulation_steps: 1
+var_train_expr: null  # trainable variable expr (eg. 'embeddings|encoder|decoder' )
+                      # must separate by |. if var_train_expr is null then we 
+                      # training all variable
+###########################################################
+#                    INTERVAL SETTING                     #
+###########################################################
+train_max_steps: 200000               # Number of training steps.
+save_interval_steps: 5000             # Interval steps to save checkpoint.
+eval_interval_steps: 500              # Interval steps to evaluate the network.
+log_interval_steps: 200               # Interval steps to record the training log.
+###########################################################
+#                     OTHER SETTING                       #
+###########################################################
+num_save_intermediate_results: 1  # Number of batch to be saved as intermediate results.

fastspeech2/model.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8be313477c9d672be6d9893dc7a6757dc6c8592425d52f133a90077f0fce20ca
+size 124798368

fastspeech2/processor.json

@@ -0,0 +1 @@
+{"symbol_to_id": {"pad": 0, "-": 1, "!": 2, "'": 3, "(": 4, ")": 5, ",": 6, ".": 7, ":": 8, ";": 9, "?": 10, " ": 11, "A": 12, "B": 13, "C": 14, "D": 15, "E": 16, "F": 17, "G": 18, "H": 19, "I": 20, "J": 21, "K": 22, "L": 23, "M": 24, "N": 25, "O": 26, "P": 27, "Q": 28, "R": 29, "S": 30, "T": 31, "U": 32, "V": 33, "W": 34, "X": 35, "Y": 36, "Z": 37, "a": 38, "b": 39, "c": 40, "d": 41, "e": 42, "f": 43, "g": 44, "h": 45, "i": 46, "j": 47, "k": 48, "l": 49, "m": 50, "n": 51, "o": 52, "p": 53, "q": 54, "r": 55, "s": 56, "t": 57, "u": 58, "v": 59, "w": 60, "x": 61, "y": 62, "z": 63, "@AA": 64, "@AA0": 65, "@AA1": 66, "@AA2": 67, "@AE": 68, "@AE0": 69, "@AE1": 70, "@AE2": 71, "@AH": 72, "@AH0": 73, "@AH1": 74, "@AH2": 75, "@AO": 76, "@AO0": 77, "@AO1": 78, "@AO2": 79, "@AW": 80, "@AW0": 81, "@AW1": 82, "@AW2": 83, "@AY": 84, "@AY0": 85, "@AY1": 86, "@AY2": 87, "@B": 88, "@CH": 89, "@D": 90, "@DH": 91, "@EH": 92, "@EH0": 93, "@EH1": 94, "@EH2": 95, "@ER": 96, "@ER0": 97, "@ER1": 98, "@ER2": 99, "@EY": 100, "@EY0": 101, "@EY1": 102, "@EY2": 103, "@F": 104, "@G": 105, "@HH": 106, "@IH": 107, "@IH0": 108, "@IH1": 109, "@IH2": 110, "@IY": 111, "@IY0": 112, "@IY1": 113, "@IY2": 114, "@JH": 115, "@K": 116, "@L": 117, "@M": 118, "@N": 119, "@NG": 120, "@OW": 121, "@OW0": 122, "@OW1": 123, "@OW2": 124, "@OY": 125, "@OY0": 126, "@OY1": 127, "@OY2": 128, "@P": 129, "@R": 130, "@S": 131, "@SH": 132, "@T": 133, "@TH": 134, "@UH": 135, "@UH0": 136, "@UH1": 137, "@UH2": 138, "@UW": 139, "@UW0": 140, "@UW1": 141, "@UW2": 142, "@V": 143, "@W": 144, "@Y": 145, "@Z": 146, "@ZH": 147, "eos": 148}, "id_to_symbol": {"0": "pad", "1": "-", "2": "!", "3": "'", "4": "(", "5": ")", "6": ",", "7": ".", "8": ":", "9": ";", "10": "?", "11": " ", "12": "A", "13": "B", "14": "C", "15": "D", "16": "E", "17": "F", "18": "G", "19": "H", "20": "I", "21": "J", "22": "K", "23": "L", "24": "M", "25": "N", "26": "O", "27": "P", "28": "Q", "29": "R", "30": "S", "31": "T", "32": "U", "33": "V", "34": "W", "35": "X", "36": "Y", "37": "Z", "38": "a", "39": "b", "40": "c", "41": "d", "42": "e", "43": "f", "44": "g", "45": "h", "46": "i", "47": "j", "48": "k", "49": "l", "50": "m", "51": "n", "52": "o", "53": "p", "54": "q", "55": "r", "56": "s", "57": "t", "58": "u", "59": "v", "60": "w", "61": "x", "62": "y", "63": "z", "64": "@AA", "65": "@AA0", "66": "@AA1", "67": "@AA2", "68": "@AE", "69": "@AE0", "70": "@AE1", "71": "@AE2", "72": "@AH", "73": "@AH0", "74": "@AH1", "75": "@AH2", "76": "@AO", "77": "@AO0", "78": "@AO1", "79": "@AO2", "80": "@AW", "81": "@AW0", "82": "@AW1", "83": "@AW2", "84": "@AY", "85": "@AY0", "86": "@AY1", "87": "@AY2", "88": "@B", "89": "@CH", "90": "@D", "91": "@DH", "92": "@EH", "93": "@EH0", "94": "@EH1", "95": "@EH2", "96": "@ER", "97": "@ER0", "98": "@ER1", "99": "@ER2", "100": "@EY", "101": "@EY0", "102": "@EY1", "103": "@EY2", "104": "@F", "105": "@G", "106": "@HH", "107": "@IH", "108": "@IH0", "109": "@IH1", "110": "@IH2", "111": "@IY", "112": "@IY0", "113": "@IY1", "114": "@IY2", "115": "@JH", "116": "@K", "117": "@L", "118": "@M", "119": "@N", "120": "@NG", "121": "@OW", "122": "@OW0", "123": "@OW1", "124": "@OW2", "125": "@OY", "126": "@OY0", "127": "@OY1", "128": "@OY2", "129": "@P", "130": "@R", "131": "@S", "132": "@SH", "133": "@T", "134": "@TH", "135": "@UH", "136": "@UH0", "137": "@UH1", "138": "@UH2", "139": "@UW", "140": "@UW0", "141": "@UW1", "142": "@UW2", "143": "@V", "144": "@W", "145": "@Y", "146": "@Z", "147": "@ZH", "148": "eos"}, "speakers_map": {"ljspeech": 0}, "processor_name": "LJSpeechProcessor"}

mb-melgan/config.yml

@@ -0,0 +1,107 @@
+
+# This is the hyperparameter configuration file for Multi-Band MelGAN.
+# Please make sure this is adjusted for the LJSpeech dataset. If you want to
+# apply to the other dataset, you might need to carefully change some parameters.
+# This configuration performs 1000k iters.
+
+###########################################################
+#                FEATURE EXTRACTION SETTING               #
+###########################################################
+sampling_rate: 22050
+hop_size: 256            # Hop size.
+format: "npy"
+
+
+###########################################################
+#         GENERATOR NETWORK ARCHITECTURE SETTING          #
+###########################################################
+model_type: "multiband_melgan_generator"
+
+multiband_melgan_generator_params:
+    out_channels: 4               # Number of output channels (number of subbands).
+    kernel_size: 7                # Kernel size of initial and final conv layers.
+    filters: 384                  # Initial number of channels for conv layers.
+    upsample_scales: [8, 4, 2]    # List of Upsampling scales.
+    stack_kernel_size: 3          # Kernel size of dilated conv layers in residual stack.
+    stacks: 4                     # Number of stacks in a single residual stack module.
+    is_weight_norm: false         # Use weight-norm or not.
+
+###########################################################
+#       DISCRIMINATOR NETWORK ARCHITECTURE SETTING        #
+###########################################################
+multiband_melgan_discriminator_params:
+    out_channels: 1                   # Number of output channels.
+    scales: 3                         # Number of multi-scales.
+    downsample_pooling: "AveragePooling1D"   # Pooling type for the input downsampling.
+    downsample_pooling_params:        # Parameters of the above pooling function.
+        pool_size: 4
+        strides: 2
+    kernel_sizes: [5, 3]              # List of kernel size.
+    filters: 16                       # Number of channels of the initial conv layer.
+    max_downsample_filters: 512       # Maximum number of channels of downsampling layers.
+    downsample_scales: [4, 4, 4]      # List of downsampling scales.
+    nonlinear_activation: "LeakyReLU" # Nonlinear activation function.
+    nonlinear_activation_params:      # Parameters of nonlinear activation function.
+        alpha: 0.2
+    is_weight_norm: false             # Use weight-norm or not.
+
+###########################################################
+#                   STFT LOSS SETTING                     #
+###########################################################
+stft_loss_params:
+    fft_lengths: [1024, 2048, 512]  # List of FFT size for STFT-based loss.
+    frame_steps: [120, 240, 50]     # List of hop size for STFT-based loss
+    frame_lengths: [600, 1200, 240] # List of window length for STFT-based loss.
+
+subband_stft_loss_params:
+    fft_lengths: [384, 683, 171]  # List of FFT size for STFT-based loss.
+    frame_steps: [30, 60, 10]     # List of hop size for STFT-based loss
+    frame_lengths: [150, 300, 60] # List of window length for STFT-based loss.
+
+###########################################################
+#               ADVERSARIAL LOSS SETTING                  #
+###########################################################
+lambda_feat_match: 10.0      # Loss balancing coefficient for feature matching loss
+lambda_adv: 2.5              # Loss balancing coefficient for adversarial loss.
+
+###########################################################
+#                  DATA LOADER SETTING                    #
+###########################################################
+batch_size: 64                 # Batch size for each GPU with assuming that gradient_accumulation_steps == 1.
+batch_max_steps: 8192          # Length of each audio in batch for training. Make sure dividable by hop_size.
+batch_max_steps_valid: 8192    # Length of each audio for validation. Make sure dividable by hope_size.
+remove_short_samples: true     # Whether to remove samples the length of which are less than batch_max_steps.
+allow_cache: true              # Whether to allow cache in dataset. If true, it requires cpu memory.
+is_shuffle: true               # shuffle dataset after each epoch.
+
+###########################################################
+#             OPTIMIZER & SCHEDULER SETTING               #
+###########################################################
+generator_optimizer_params:
+    lr_fn: "PiecewiseConstantDecay"
+    lr_params: 
+        boundaries: [100000, 200000, 300000, 400000, 500000, 600000, 700000]
+        values: [0.0005, 0.0005, 0.00025, 0.000125, 0.0000625, 0.00003125, 0.000015625, 0.000001]
+    amsgrad: false
+
+discriminator_optimizer_params:
+    lr_fn: "PiecewiseConstantDecay"
+    lr_params: 
+        boundaries: [100000, 200000, 300000, 400000, 500000]
+        values: [0.00025, 0.000125, 0.0000625, 0.00003125, 0.000015625, 0.000001]
+    amsgrad: false
+
+gradient_accumulation_steps: 1
+###########################################################
+#                    INTERVAL SETTING                     #
+###########################################################
+discriminator_train_start_steps: 200000  # steps begin training discriminator
+train_max_steps: 4000000                 # Number of training steps.
+save_interval_steps: 20000               # Interval steps to save checkpoint.
+eval_interval_steps: 5000                # Interval steps to evaluate the network.
+log_interval_steps: 200                  # Interval steps to record the training log.
+
+###########################################################
+#                     OTHER SETTING                       #
+###########################################################
+num_save_intermediate_results: 1  # Number of batch to be saved as intermediate results.

mb-melgan/model.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b53fe8116ec61fb7a630377f120ef8ff8fa1e8b1ca5f9756259fe3e1e304ef68
+size 10234448

melgan-stft/melgan.stft-2M.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f7e40f5c67694fe02a7c96d293203855a9e261442bf721f46089e384b80b45b5
+size 17142440

melgan-stft/melgan.stft_config.yml

@@ -0,0 +1,89 @@
+allow_cache: true
+batch_max_steps: 8192
+batch_max_steps_valid: 81920
+batch_size: 16
+config: egs/ljspeech/conf/multi_stft_melgan.yaml
+dev_dir: ./egs/ljspeech/dump/valid
+discriminator_mixed_precision: false
+discriminator_optimizer_params:
+  beta_1: 0.5
+  beta_2: 0.9
+  decay_steps: 4000000
+  end_learning_rate: 0.0001
+  initial_learning_rate: 0.0001
+  warmup_steps: 0
+  weight_decay: 0.0
+discriminator_params:
+  downsample_pooling: AveragePooling1D
+  downsample_pooling_params:
+    pool_size: 4
+    strides: 2
+  downsample_scales:
+  - 4
+  - 4
+  - 4
+  - 4
+  filters: 16
+  is_weight_norm: false
+  kernel_sizes:
+  - 5
+  - 3
+  max_downsample_filters: 1024
+  nonlinear_activation: LeakyReLU
+  nonlinear_activation_params:
+    alpha: 0.2
+  out_channels: 1
+  scales: 3
+discriminator_train_start_steps: 100000
+eval_interval_steps: 5000
+format: npy
+generator_mixed_precision: false
+generator_optimizer_params:
+  beta_1: 0.5
+  beta_2: 0.9
+  decay_steps: 100000
+  end_learning_rate: 0.0001
+  initial_learning_rate: 0.001
+  warmup_steps: 0
+  weight_decay: 0.0
+generator_params:
+  filters: 512
+  is_weight_norm: false
+  kernel_size: 7
+  out_channels: 1
+  stack_kernel_size: 3
+  stacks: 3
+  upsample_scales:
+  - 8
+  - 8
+  - 2
+  - 2
+hop_size: 256
+is_shuffle: true
+lambda_adv: 4.0
+lambda_feat_match: 10.0
+log_interval_steps: 200
+num_save_intermediate_results: 1
+outdir: ./egs/ljspeech/exp/melgan.stft.v2/
+remove_short_samples: true
+resume: egs/ljspeech/exp/melgan.stft.v2/checkpoints/ckpt-100000
+sampling_rate: 22050
+save_interval_steps: 20000
+stft_loss_params:
+  fft_lengths:
+  - 1024
+  - 2048
+  - 512
+  frame_lengths:
+  - 600
+  - 1200
+  - 240
+  frame_steps:
+  - 120
+  - 240
+  - 50
+train_dir: ./egs/ljspeech/dump/train/
+train_max_steps: 4000000
+use_norm: true
+verbose: 1
+version: 0.3.4

melgan-stft/melgan_stft.v1.yaml

@@ -0,0 +1,102 @@
+
+# This is the hyperparameter configuration file for MelGAN with Multi Resolution STFT.
+# Please make sure this is adjusted for the LJSpeech dataset. If you want to
+# apply to the other dataset, you might need to carefully change some parameters.
+# This configuration performs 4000k iters.
+
+###########################################################
+#                FEATURE EXTRACTION SETTING               #
+###########################################################
+sampling_rate: 22050
+hop_size: 256            # Hop size.
+format: "npy"
+
+
+###########################################################
+#         GENERATOR NETWORK ARCHITECTURE SETTING          #
+###########################################################
+model_type: "melgan_generator"
+
+melgan_generator_params:
+    out_channels: 1               # Number of output channels.
+    kernel_size: 7                # Kernel size of initial and final conv layers.
+    filters: 512                  # Initial number of channels for conv layers.
+    upsample_scales: [8, 8, 2, 2] # List of Upsampling scales.
+    stack_kernel_size: 3          # Kernel size of dilated conv layers in residual stack.
+    stacks: 3                     # Number of stacks in a single residual stack module.
+    is_weight_norm: false
+
+###########################################################
+#       DISCRIMINATOR NETWORK ARCHITECTURE SETTING        #
+###########################################################
+melgan_discriminator_params:
+    out_channels: 1                   # Number of output channels.
+    scales: 3                         # Number of multi-scales.
+    downsample_pooling: "AveragePooling1D"   # Pooling type for the input downsampling.
+    downsample_pooling_params:        # Parameters of the above pooling function.
+        pool_size: 4
+        strides: 2
+    kernel_sizes: [5, 3]              # List of kernel size.
+    filters: 16                       # Number of channels of the initial conv layer.
+    max_downsample_filters: 1024      # Maximum number of channels of downsampling layers.
+    downsample_scales: [4, 4, 4, 4]   # List of downsampling scales.
+    nonlinear_activation: "LeakyReLU" # Nonlinear activation function.
+    nonlinear_activation_params:      # Parameters of nonlinear activation function.
+        alpha: 0.2
+    is_weight_norm: false
+
+###########################################################
+#                   STFT LOSS SETTING                     #
+###########################################################
+stft_loss_params:
+    fft_lengths: [1024, 2048, 512]  # List of FFT size for STFT-based loss.
+    frame_steps: [120, 240, 50]     # List of hop size for STFT-based loss
+    frame_lengths: [600, 1200, 240] # List of window length for STFT-based loss.
+
+
+###########################################################
+#               ADVERSARIAL LOSS SETTING                  #
+###########################################################
+lambda_feat_match: 10.0
+lambda_adv: 4.0
+
+###########################################################
+#                  DATA LOADER SETTING                    #
+###########################################################
+batch_size: 16                 # Batch size for each GPU with assuming that gradient_accumulation_steps == 1.
+batch_max_steps: 8192          # Length of each audio in batch for training. Make sure dividable by hop_size.
+batch_max_steps_valid: 81920   # Length of each audio for validation. Make sure dividable by hope_size.
+remove_short_samples: true     # Whether to remove samples the length of which are less than batch_max_steps.
+allow_cache: true              # Whether to allow cache in dataset. If true, it requires cpu memory.
+is_shuffle: true               # shuffle dataset after each epoch.
+
+###########################################################
+#             OPTIMIZER & SCHEDULER SETTING               #
+###########################################################
+generator_optimizer_params:
+    lr_fn: "PiecewiseConstantDecay"
+    lr_params: 
+        boundaries: [100000]       # = discriminator_train_start_steps.
+        values: [0.0005, 0.0001]    # learning rate each interval.
+
+    
+discriminator_optimizer_params:
+    lr_fn: "PiecewiseConstantDecay"
+    lr_params: 
+        boundaries: [0]             # after resume and start training discriminator, global steps is 100k, but local discriminator step is 0
+        values: [0.0001, 0.0001]    # learning rate each interval.
+
+gradient_accumulation_steps: 1
+###########################################################
+#                    INTERVAL SETTING                     #
+###########################################################
+discriminator_train_start_steps: 100000  # steps begin training discriminator
+train_max_steps: 4000000                 # Number of training steps.
+save_interval_steps: 20000               # Interval steps to save checkpoint.
+eval_interval_steps: 5000                # Interval steps to evaluate the network.
+log_interval_steps: 200                  # Interval steps to record the training log.
+
+###########################################################
+#                     OTHER SETTING                       #
+###########################################################
+num_save_intermediate_results: 1  # Number of batch to be saved as intermediate results.

melgan/config.yml

@@ -0,0 +1,89 @@
+
+# This is the hyperparameter configuration file for MelGAN.
+# Please make sure this is adjusted for the LJSpeech dataset. If you want to
+# apply to the other dataset, you might need to carefully change some parameters.
+# This configuration performs 4000k iters.
+
+###########################################################
+#                FEATURE EXTRACTION SETTING               #
+###########################################################
+sampling_rate: 22050     # Sampling rate of dataset.
+hop_size: 256            # Hop size.
+format: "npy"
+
+
+###########################################################
+#         GENERATOR NETWORK ARCHITECTURE SETTING          #
+###########################################################
+model_type: "melgan_generator"
+
+melgan_generator_params:
+    out_channels: 1               # Number of output channels.
+    kernel_size: 7                # Kernel size of initial and final conv layers.
+    filters: 512                  # Initial number of channels for conv layers.
+    upsample_scales: [8, 8, 2, 2] # List of Upsampling scales.
+    stack_kernel_size: 3          # Kernel size of dilated conv layers in residual stack.
+    stacks: 3                     # Number of stacks in a single residual stack module.
+    is_weight_norm: false         # Use weight-norm or not.
+
+###########################################################
+#       DISCRIMINATOR NETWORK ARCHITECTURE SETTING        #
+###########################################################
+melgan_discriminator_params:
+    out_channels: 1                   # Number of output channels.
+    scales: 3                         # Number of multi-scales.
+    downsample_pooling: "AveragePooling1D"   # Pooling type for the input downsampling.
+    downsample_pooling_params:        # Parameters of the above pooling function.
+        pool_size: 4
+        strides: 2
+    kernel_sizes: [5, 3]              # List of kernel size.
+    filters: 16                       # Number of channels of the initial conv layer.
+    max_downsample_filters: 1024      # Maximum number of channels of downsampling layers.
+    downsample_scales: [4, 4, 4, 4]   # List of downsampling scales.
+    nonlinear_activation: "LeakyReLU" # Nonlinear activation function.
+    nonlinear_activation_params:      # Parameters of nonlinear activation function.
+        alpha: 0.2
+    is_weight_norm: false             # Use weight-norm or not.
+
+###########################################################
+#               ADVERSARIAL LOSS SETTING                  #
+###########################################################
+lambda_feat_match: 10.0
+
+###########################################################
+#                  DATA LOADER SETTING                    #
+###########################################################
+batch_size: 16                 # Batch size for each GPU with assuming that gradient_accumulation_steps == 1.
+batch_max_steps: 8192          # Length of each audio in batch for training. Make sure dividable by hop_size.
+batch_max_steps_valid: 81920   # Length of each audio for validation. Make sure dividable by hope_size.
+remove_short_samples: true     # Whether to remove samples the length of which are less than batch_max_steps.
+allow_cache: true              # Whether to allow cache in dataset. If true, it requires cpu memory.
+is_shuffle: true               # shuffle dataset after each epoch.
+
+###########################################################
+#             OPTIMIZER & SCHEDULER SETTING               #
+###########################################################
+generator_optimizer_params:
+    lr: 0.0001             # Generator's learning rate.
+    beta_1: 0.5
+    beta_2: 0.9
+    
+discriminator_optimizer_params:
+    lr: 0.0001            # Discriminator's learning rate.
+    beta_1: 0.5
+    beta_2: 0.9
+
+gradient_accumulation_steps: 1
+###########################################################
+#                    INTERVAL SETTING                     #
+###########################################################
+train_max_steps: 4000000                # Number of training steps.
+save_interval_steps: 3               # Interval steps to save checkpoint.
+eval_interval_steps: 2              # Interval steps to evaluate the network.
+log_interval_steps: 1                # Interval steps to record the training log.
+discriminator_train_start_steps: 0     # step to start training discriminator.
+
+###########################################################
+#                     OTHER SETTING                       #
+###########################################################
+num_save_intermediate_results: 1  # Number of batch to be saved as intermediate results.

melgan/model.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:33204255555aa9992443585cfdff3befe88d2f60b94224f0493acdf43ee84f19
+size 17142440

tacatron2/config.yml

@@ -0,0 +1,86 @@
+# This is the hyperparameter configuration file for Tacotron2 v1.
+# Please make sure this is adjusted for the LJSpeech dataset. If you want to
+# apply to the other dataset, you might need to carefully change some parameters.
+# This configuration performs 200k iters but 65k iters is enough to get a good models.
+
+###########################################################
+#                FEATURE EXTRACTION SETTING               #
+###########################################################
+hop_size: 256            # Hop size.
+format: "npy"
+
+
+###########################################################
+#              NETWORK ARCHITECTURE SETTING               #
+###########################################################
+model_type: "tacotron2"
+
+tacotron2_params:
+    dataset: ljspeech
+    embedding_hidden_size: 512
+    initializer_range: 0.02
+    embedding_dropout_prob: 0.1
+    n_speakers: 1
+    n_conv_encoder: 5
+    encoder_conv_filters: 512
+    encoder_conv_kernel_sizes: 5
+    encoder_conv_activation: 'relu'
+    encoder_conv_dropout_rate: 0.5
+    encoder_lstm_units: 256
+    n_prenet_layers: 2
+    prenet_units: 256
+    prenet_activation: 'relu'
+    prenet_dropout_rate: 0.5
+    n_lstm_decoder: 1
+    reduction_factor: 1
+    decoder_lstm_units: 1024
+    attention_dim: 128
+    attention_filters: 32
+    attention_kernel: 31
+    n_mels: 80
+    n_conv_postnet: 5
+    postnet_conv_filters: 512
+    postnet_conv_kernel_sizes: 5
+    postnet_dropout_rate: 0.1
+    attention_type: "lsa"
+
+###########################################################
+#                  DATA LOADER SETTING                    #
+###########################################################
+batch_size: 32             # Batch size for each GPU with assuming that gradient_accumulation_steps == 1.
+remove_short_samples: true # Whether to remove samples the length of which are less than batch_max_steps.
+allow_cache: true          # Whether to allow cache in dataset. If true, it requires cpu memory.
+mel_length_threshold: 32   # remove all targets has mel_length <= 32 
+is_shuffle: true           # shuffle dataset after each epoch.
+use_fixed_shapes: true     # use_fixed_shapes for training (2x speed-up)
+                           # refer (https://github.com/dathudeptrai/TensorflowTTS/issues/34#issuecomment-642309118)
+
+###########################################################
+#             OPTIMIZER & SCHEDULER SETTING               #
+###########################################################
+optimizer_params:
+    initial_learning_rate: 0.001
+    end_learning_rate: 0.00001
+    decay_steps: 150000          # < train_max_steps is recommend.
+    warmup_proportion: 0.02
+    weight_decay: 0.001
+    
+gradient_accumulation_steps: 1
+var_train_expr: null  # trainable variable expr (eg. 'embeddings|decoder_cell' )
+                      # must separate by |. if var_train_expr is null then we 
+                      # training all variables.
+###########################################################
+#                    INTERVAL SETTING                     #
+###########################################################
+train_max_steps: 200000                 # Number of training steps.
+save_interval_steps: 2000               # Interval steps to save checkpoint.
+eval_interval_steps: 500                # Interval steps to evaluate the network.
+log_interval_steps: 200                 # Interval steps to record the training log.
+start_schedule_teacher_forcing: 200001  # don't need to apply schedule teacher forcing.
+start_ratio_value: 0.5                  # start ratio of scheduled teacher forcing.
+schedule_decay_steps: 50000             # decay step scheduled teacher forcing.
+end_ratio_value: 0.0                    # end ratio of scheduled teacher forcing.
+###########################################################
+#                     OTHER SETTING                       #
+###########################################################
+num_save_intermediate_results: 1  # Number of results to be saved as intermediate results.

tacatron2/model.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5f1a59e2e0fb31325b9103762f8709fa8f1f90f3fb8fc095d4dabb7e6722d406
+size 127975304

tacatron2/processor.json

@@ -0,0 +1 @@
+{"symbol_to_id": {"pad": 0, "-": 1, "!": 2, "'": 3, "(": 4, ")": 5, ",": 6, ".": 7, ":": 8, ";": 9, "?": 10, " ": 11, "A": 12, "B": 13, "C": 14, "D": 15, "E": 16, "F": 17, "G": 18, "H": 19, "I": 20, "J": 21, "K": 22, "L": 23, "M": 24, "N": 25, "O": 26, "P": 27, "Q": 28, "R": 29, "S": 30, "T": 31, "U": 32, "V": 33, "W": 34, "X": 35, "Y": 36, "Z": 37, "a": 38, "b": 39, "c": 40, "d": 41, "e": 42, "f": 43, "g": 44, "h": 45, "i": 46, "j": 47, "k": 48, "l": 49, "m": 50, "n": 51, "o": 52, "p": 53, "q": 54, "r": 55, "s": 56, "t": 57, "u": 58, "v": 59, "w": 60, "x": 61, "y": 62, "z": 63, "@AA": 64, "@AA0": 65, "@AA1": 66, "@AA2": 67, "@AE": 68, "@AE0": 69, "@AE1": 70, "@AE2": 71, "@AH": 72, "@AH0": 73, "@AH1": 74, "@AH2": 75, "@AO": 76, "@AO0": 77, "@AO1": 78, "@AO2": 79, "@AW": 80, "@AW0": 81, "@AW1": 82, "@AW2": 83, "@AY": 84, "@AY0": 85, "@AY1": 86, "@AY2": 87, "@B": 88, "@CH": 89, "@D": 90, "@DH": 91, "@EH": 92, "@EH0": 93, "@EH1": 94, "@EH2": 95, "@ER": 96, "@ER0": 97, "@ER1": 98, "@ER2": 99, "@EY": 100, "@EY0": 101, "@EY1": 102, "@EY2": 103, "@F": 104, "@G": 105, "@HH": 106, "@IH": 107, "@IH0": 108, "@IH1": 109, "@IH2": 110, "@IY": 111, "@IY0": 112, "@IY1": 113, "@IY2": 114, "@JH": 115, "@K": 116, "@L": 117, "@M": 118, "@N": 119, "@NG": 120, "@OW": 121, "@OW0": 122, "@OW1": 123, "@OW2": 124, "@OY": 125, "@OY0": 126, "@OY1": 127, "@OY2": 128, "@P": 129, "@R": 130, "@S": 131, "@SH": 132, "@T": 133, "@TH": 134, "@UH": 135, "@UH0": 136, "@UH1": 137, "@UH2": 138, "@UW": 139, "@UW0": 140, "@UW1": 141, "@UW2": 142, "@V": 143, "@W": 144, "@Y": 145, "@Z": 146, "@ZH": 147, "eos": 148}, "id_to_symbol": {"0": "pad", "1": "-", "2": "!", "3": "'", "4": "(", "5": ")", "6": ",", "7": ".", "8": ":", "9": ";", "10": "?", "11": " ", "12": "A", "13": "B", "14": "C", "15": "D", "16": "E", "17": "F", "18": "G", "19": "H", "20": "I", "21": "J", "22": "K", "23": "L", "24": "M", "25": "N", "26": "O", "27": "P", "28": "Q", "29": "R", "30": "S", "31": "T", "32": "U", "33": "V", "34": "W", "35": "X", "36": "Y", "37": "Z", "38": "a", "39": "b", "40": "c", "41": "d", "42": "e", "43": "f", "44": "g", "45": "h", "46": "i", "47": "j", "48": "k", "49": "l", "50": "m", "51": "n", "52": "o", "53": "p", "54": "q", "55": "r", "56": "s", "57": "t", "58": "u", "59": "v", "60": "w", "61": "x", "62": "y", "63": "z", "64": "@AA", "65": "@AA0", "66": "@AA1", "67": "@AA2", "68": "@AE", "69": "@AE0", "70": "@AE1", "71": "@AE2", "72": "@AH", "73": "@AH0", "74": "@AH1", "75": "@AH2", "76": "@AO", "77": "@AO0", "78": "@AO1", "79": "@AO2", "80": "@AW", "81": "@AW0", "82": "@AW1", "83": "@AW2", "84": "@AY", "85": "@AY0", "86": "@AY1", "87": "@AY2", "88": "@B", "89": "@CH", "90": "@D", "91": "@DH", "92": "@EH", "93": "@EH0", "94": "@EH1", "95": "@EH2", "96": "@ER", "97": "@ER0", "98": "@ER1", "99": "@ER2", "100": "@EY", "101": "@EY0", "102": "@EY1", "103": "@EY2", "104": "@F", "105": "@G", "106": "@HH", "107": "@IH", "108": "@IH0", "109": "@IH1", "110": "@IH2", "111": "@IY", "112": "@IY0", "113": "@IY1", "114": "@IY2", "115": "@JH", "116": "@K", "117": "@L", "118": "@M", "119": "@N", "120": "@NG", "121": "@OW", "122": "@OW0", "123": "@OW1", "124": "@OW2", "125": "@OY", "126": "@OY0", "127": "@OY1", "128": "@OY2", "129": "@P", "130": "@R", "131": "@S", "132": "@SH", "133": "@T", "134": "@TH", "135": "@UH", "136": "@UH0", "137": "@UH1", "138": "@UH2", "139": "@UW", "140": "@UW0", "141": "@UW1", "142": "@UW2", "143": "@V", "144": "@W", "145": "@Y", "146": "@Z", "147": "@ZH", "148": "eos"}, "speakers_map": {"ljspeech": 0}, "processor_name": "LJSpeechProcessor"}