End of training

README.md

@@ -3,20 +3,25 @@ library_name: transformers
 language:
 - en
 license: apache-2.0
-base_model: openai/whisper-tiny.en
+base_model: openai/whisper-medium.en
 tags:
 - generated_from_trainer
+metrics:
+- wer
 model-index:
-- name: articuplay
+- name: articuplay1
   results: []
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 should probably proofread and complete it, then remove this comment. -->
 
-# articuplay
+# articuplay1
 
-This model is a fine-tuned version of [openai/whisper-tiny.en](https://huggingface.co/openai/whisper-tiny.en) on the TIMIT dataset.
+This model is a fine-tuned version of [openai/whisper-medium.en](https://huggingface.co/openai/whisper-medium.en) on the TIMIT dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.2033
+- Wer: 13.2533
 
 ## Model description
 
@@ -42,12 +47,23 @@ The following hyperparameters were used during training:
 - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_steps: 500
-- num_epochs: 10
+- num_epochs: 5
 - mixed_precision_training: Native AMP
 
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Wer     |
+|:-------------:|:-----:|:----:|:---------------:|:-------:|
+| 2.9702        | 1.0   | 289  | 2.4632          | 95.9398 |
+| 0.2224        | 2.0   | 578  | 0.2168          | 14.6648 |
+| 0.149         | 3.0   | 867  | 0.1952          | 13.7123 |
+| 0.1087        | 4.0   | 1156 | 0.1945          | 13.2154 |
+| 0.0763        | 5.0   | 1445 | 0.2033          | 13.2533 |
+
+
 ### Framework versions
 
-- Transformers 4.49.0
+- Transformers 4.51.3
 - Pytorch 2.6.0+cu124
 - Datasets 3.1.0
 - Tokenizers 0.21.0

config.json

@@ -1,5 +1,4 @@
 {
-  "_name_or_path": "openai/whisper-tiny.en",
   "activation_dropout": 0.0,
   "activation_function": "gelu",
   "apply_spec_augment": false,
@@ -13,24 +12,19 @@
   ],
   "bos_token_id": 50257,
   "classifier_proj_size": 256,
-  "d_model": 384,
-  "decoder_attention_heads": 6,
-  "decoder_ffn_dim": 1536,
+  "d_model": 768,
+  "decoder_attention_heads": 12,
+  "decoder_ffn_dim": 3072,
   "decoder_layerdrop": 0.0,
-  "decoder_layers": 4,
+  "decoder_layers": 12,
   "decoder_start_token_id": 50257,
   "dropout": 0.0,
-  "encoder_attention_heads": 6,
-  "encoder_ffn_dim": 1536,
+  "encoder_attention_heads": 12,
+  "encoder_ffn_dim": 3072,
   "encoder_layerdrop": 0.0,
-  "encoder_layers": 4,
+  "encoder_layers": 12,
   "eos_token_id": 50256,
-  "forced_decoder_ids": [
-    [
-      1,
-      50362
-    ]
-  ],
+  "forced_decoder_ids": null,
   "init_std": 0.02,
   "is_encoder_decoder": true,
   "mask_feature_length": 10,
@@ -44,12 +38,12 @@
   "max_target_positions": 448,
   "median_filter_width": 7,
   "model_type": "whisper",
-  "num_hidden_layers": 4,
+  "num_hidden_layers": 12,
   "num_mel_bins": 80,
   "pad_token_id": 50256,
   "scale_embedding": false,
   "torch_dtype": "float32",
-  "transformers_version": "4.49.0",
+  "transformers_version": "4.51.3",
   "use_cache": true,
   "use_weighted_layer_sum": false,
   "vocab_size": 51880

generation_config.json

@@ -6,12 +6,6 @@
   "bos_token_id": 50257,
   "decoder_start_token_id": 50257,
   "eos_token_id": 50256,
-  "forced_decoder_ids": [
-    [
-      1,
-      50362
-    ]
-  ],
   "max_length": 448,
   "pad_token_id": 50256,
   "suppress_tokens": [
@@ -106,5 +100,5 @@
     50360,
     50361
   ],
-  "transformers_version": "4.49.0"
+  "transformers_version": "4.51.3"
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:17d7274cd97799226121d5b7c1c57206e746b618c3e4c8127beb9ee86a22e3cb
-size 151084712
+oid sha256:ce831ec91bceda2ccc7e3de9b2b4f8d9a9426e58d10f2093059900f931a50479
+size 967041160

preprocessor_config.json

@@ -1,5 +1,6 @@
 {
   "chunk_length": 30,
+  "dither": 0.0,
   "feature_extractor_type": "WhisperFeatureExtractor",
   "feature_size": 80,
   "hop_length": 160,

runs/May07_16-10-41_bonestew/events.out.tfevents.1746648642.bonestew.31073.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:849aa0d9200811219d9acad1c8ed6c69e67049fdca4b36075f6ef66102b2246b
+size 6734

runs/May07_16-22-44_bonestew/events.out.tfevents.1746649364.bonestew.769.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9aae7127ea7a7ba46c56eefcb0fce52e0349b2e789b49771fb3a26a0fa638b39
+size 7148

runs/May07_16-40-30_bonestew/events.out.tfevents.1746650431.bonestew.769.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ad7cbb440fed9de9290799dce2c23322ec31ce24d0118246ab45d89aa2a97d9d
+size 6940

runs/May07_16-54-33_bonestew/events.out.tfevents.1746651274.bonestew.769.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:601c79153f141b4e2076d5b92f964d6737d9fda3472cee766658e44a046b5a00
+size 12593

runs/May07_17-23-08_bonestew/events.out.tfevents.1746652989.bonestew.769.3

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:36b4bc26ce44fcc9184eaa24868dc9ccffd9e54810977cacf073fdc3a9361bac
+size 99723

runs/May07_20-16-13_bonestew/events.out.tfevents.1746663374.bonestew.769.4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f92e3ca89df8cc8462580870010d9c4ba1f2909a4aa6edce063560959b7d7036
+size 38978

runs/May07_22-48-36_bonestew/events.out.tfevents.1746672535.bonestew.769.5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2e9b10e70a9932ba25d1a1cb6f9cbc17833a055d47637ed6c234a332063b724a
+size 6906

runs/May07_23-19-57_bonestew/events.out.tfevents.1746674401.bonestew.117091.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:41d1bbd05be6fa83c3fd870a1c53b55d93bf2420f2cd48806b92df732dc2448c
+size 38978

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:895b92d3b49beb2d333b2040c31456ae21521fd37389a99a456cf78ea51a1cb8
+oid sha256:81baed9df131b1146c0ddb5c0a4c89601045fbc25c18b44724211c10cc7b0dea
 size 5496