End of training

README.md

@@ -1,8 +1,10 @@
 ---
 license: apache-2.0
-base_model: openai/whisper-small
+base_model: openai/whisper-medium
 tags:
 - generated_from_trainer
+metrics:
+- wer
 model-index:
 - name: my_whisper
   results: []
@@ -13,7 +15,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # my_whisper
 
-This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on an unknown dataset.
+This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.0000
+- Wer: 0.0
 
 ## Model description
 
@@ -39,11 +44,16 @@ The following hyperparameters were used during training:
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_steps: 3
-- training_steps: 5
+- training_steps: 15
 - mixed_precision_training: Native AMP
 
 ### Training results
 
+| Training Loss | Epoch | Step | Validation Loss | Wer  |
+|:-------------:|:-----:|:----:|:---------------:|:----:|
+| 1.2537        | 5.0   | 5    | 1.2684          | 62.5 |
+| 0.2765        | 10.0  | 10   | 0.0001          | 0.0  |
+| 0.0001        | 15.0  | 15   | 0.0000          | 0.0  |
 
 
 ### Framework versions

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "openai/whisper-large",
+  "_name_or_path": "openai/whisper-medium",
   "activation_dropout": 0.0,
   "activation_function": "gelu",
   "apply_spec_augment": false,
@@ -13,17 +13,17 @@
   ],
   "bos_token_id": 50257,
   "classifier_proj_size": 256,
-  "d_model": 1280,
-  "decoder_attention_heads": 20,
-  "decoder_ffn_dim": 5120,
+  "d_model": 1024,
+  "decoder_attention_heads": 16,
+  "decoder_ffn_dim": 4096,
   "decoder_layerdrop": 0.0,
-  "decoder_layers": 32,
+  "decoder_layers": 24,
   "decoder_start_token_id": 50258,
   "dropout": 0.0,
-  "encoder_attention_heads": 20,
-  "encoder_ffn_dim": 5120,
+  "encoder_attention_heads": 16,
+  "encoder_ffn_dim": 4096,
   "encoder_layerdrop": 0.0,
-  "encoder_layers": 32,
+  "encoder_layers": 24,
   "eos_token_id": 50257,
   "forced_decoder_ids": [
     [
@@ -52,7 +52,7 @@
   "max_target_positions": 448,
   "median_filter_width": 7,
   "model_type": "whisper",
-  "num_hidden_layers": 32,
+  "num_hidden_layers": 24,
   "num_mel_bins": 80,
   "pad_token_id": 50257,
   "scale_embedding": false,

generation_config.json

@@ -1,44 +1,28 @@
 {
   "alignment_heads": [
     [
-      5,
-      3
+      13,
+      15
     ],
     [
-      5,
-      9
-    ],
-    [
-      8,
-      0
-    ],
-    [
-      8,
+      15,
       4
     ],
     [
-      8,
-      7
+      15,
+      15
     ],
     [
-      8,
-      8
+      16,
+      1
     ],
     [
-      9,
+      20,
       0
     ],
     [
-      9,
-      7
-    ],
-    [
-      9,
-      9
-    ],
-    [
-      10,
-      5
+      23,
+      4
     ]
   ],
   "begin_suppress_tokens": [

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:156f00fff58c90080691e8bbf094e53e3190b172b0a999d187629179b640916e
-size 966995080
+oid sha256:4cff82a251042833c9a60bf6703b27e6a9196dad4d0216236fb648cc2bf693c4
+size 3055544304

runs/Jun06_18-02-39_archlinux-ai/events.out.tfevents.1717686160.archlinux-ai.44913.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d66be74adf84cee44f739a2feacb08f416a40555b5437463f67aa628296f1743
+size 6574

runs/Jun07_08-18-44_archlinux-ai/events.out.tfevents.1717737525.archlinux-ai.3960.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8bd639c8235a4cfd3e91bb5e3a55f144eecfc67bf7a96086d579518987644175
+size 6574

runs/Jun07_08-30-31_archlinux-ai/events.out.tfevents.1717738232.archlinux-ai.5016.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a330273e4ab9e797e16865e5fba1473bf82333f4ad87da7b33a201cfbc935344
+size 6574

runs/Jun07_08-34-50_archlinux-ai/events.out.tfevents.1717738492.archlinux-ai.5334.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f097d44a6333c0c17bdd9cb9b135b4d31ce19cbb34b502ca9b9d1395283aea9e
+size 6574

runs/Jun07_08-42-33_archlinux-ai/events.out.tfevents.1717738954.archlinux-ai.5508.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a48e5c0b168e40e382c67a57ba5cbc34588e956172f4e1a5615f2b2a22051af1
+size 8480

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:384c5594ecddb6c4e9b0f184ef9c41af6cae3ff3186ab5a553b3b42962108d9b
+oid sha256:32aadc17e7ef1cd761d328e05f2755cb413a03fd67986d9f7d75ad3743761b34
 size 5240