Upload SpeechT5ForSpeechToText

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "/kaggle/input/speecht5-infe/model_checkpoint/checkpoint-3000",
+  "_name_or_path": "averaged_model",
   "activation_dropout": 0.064,
   "apply_spec_augment": true,
   "architectures": [
@@ -39,40 +39,40 @@
   "decoder_ffn_dim": 3072,
   "decoder_layerdrop": 0.1,
   "decoder_layers": 6,
-  "decoder_start_token_id": 0,
+  "decoder_start_token_id": 1,
   "encoder_attention_heads": 12,
   "encoder_ffn_dim": 3072,
   "encoder_layerdrop": 0.1,
   "encoder_layers": 12,
   "encoder_max_relative_position": 160,
   "eos_token_id": 2,
-  "feat_extract_activation": "gelu",
+  "feat_extract_activation": "gelu_new",
   "feat_extract_norm": "group",
-  "feat_proj_dropout": 0.0,
+  "feat_proj_dropout": 0.08,
   "guided_attention_loss_num_heads": 2,
   "guided_attention_loss_scale": 10.0,
   "guided_attention_loss_sigma": 0.4,
   "hidden_act": "gelu_new",
-  "hidden_dropout": 0.095,
+  "hidden_dropout": 0.098,
   "hidden_size": 768,
   "initializer_range": 0.02,
   "is_encoder_decoder": true,
   "layer_norm_eps": 1e-05,
-  "mask_feature_length": 10,
-  "mask_feature_min_masks": 0,
-  "mask_feature_prob": 0.0,
-  "mask_time_length": 10,
+  "mask_feature_length": 15,
+  "mask_feature_min_masks": 1,
+  "mask_feature_prob": 0.05,
+  "mask_time_length": 7,
   "mask_time_min_masks": 2,
-  "mask_time_prob": 0.05,
+  "mask_time_prob": 0.051,
   "max_length": 450,
-  "max_speech_positions": 4000,
-  "max_text_positions": 450,
+  "max_speech_positions": 4001,
+  "max_text_positions": 451,
   "model_type": "speecht5",
   "num_conv_pos_embedding_groups": 16,
   "num_conv_pos_embeddings": 128,
   "num_feat_extract_layers": 7,
   "num_mel_bins": 80,
-  "pad_token_id": 1,
+  "pad_token_id": 0,
   "positional_dropout": 0.1,
   "reduction_factor": 2,
   "scale_embedding": true,
@@ -86,7 +86,7 @@
   "speech_decoder_prenet_units": 256,
   "torch_dtype": "float32",
   "transformers_version": "4.33.0",
-  "use_cache": true,
-  "use_guided_attention_loss": true,
-  "vocab_size": 81
+  "use_cache": false,
+  "use_guided_attention_loss": false,
+  "vocab_size": 500
 }

generation_config.json

@@ -1,9 +1,10 @@
 {
   "_from_model_config": true,
   "bos_token_id": 0,
-  "decoder_start_token_id": 2,
+  "decoder_start_token_id": 1,
   "eos_token_id": 2,
   "max_length": 450,
-  "pad_token_id": 1,
-  "transformers_version": "4.33.0"
+  "pad_token_id": 0,
+  "transformers_version": "4.33.0",
+  "use_cache": false
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5652b9ff493bcd1c6d6276ff780532ffb6ee0713d05d32600f3d199f05f7e13f
-size 618484633
+oid sha256:9c8839285f894735317705d3bcef0970e56a53722034251b1820c77523ddd2a5
+size 619777049