End of training

README.md

@@ -11,7 +11,7 @@ datasets:
 metrics:
 - wer
 model-index:
-- name: Whisper Small Ko - CookieMoster99
+- name: Whisper Small KR - CookieMoster99
   results:
   - task:
       name: Automatic Speech Recognition
@@ -19,22 +19,22 @@ model-index:
     dataset:
       name: zeroth-korean
       type: Bingsu/zeroth-korean
-      args: 'config: ko, split: test'
+      args: 'config: KR, split: test'
     metrics:
     - name: Wer
       type: wer
-      value: 86.67369372082517
+      value: 52.3565728053004
 ---
 
 <!-- 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. -->
 
-# Whisper Small Ko - CookieMoster99
+# Whisper Small KR - CookieMoster99
 
 This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the zeroth-korean dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.0884
-- Wer: 86.6737
+- Wer: 52.3566
 
 ## Model description
 
@@ -54,22 +54,21 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 1e-05
-- train_batch_size: 16
+- train_batch_size: 8
 - eval_batch_size: 8
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_steps: 500
-- training_steps: 4000
+- training_steps: 3000
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss | Wer      |
-|:-------------:|:-----:|:----:|:---------------:|:--------:|
-| 0.1053        | 0.72  | 1000 | 0.1508          | 68.2729  |
-| 0.0518        | 1.44  | 2000 | 0.1074          | 80.8613  |
-| 0.0136        | 2.16  | 3000 | 0.0918          | 106.6707 |
-| 0.013         | 2.87  | 4000 | 0.0884          | 86.6737  |
+| Training Loss | Epoch | Step | Validation Loss | Wer     |
+|:-------------:|:-----:|:----:|:---------------:|:-------:|
+| 0.0254        | 0.36  | 1000 | 0.1079          | 35.4314 |
+| 0.0141        | 0.72  | 2000 | 0.0955          | 41.0029 |
+| 0.0097        | 1.08  | 3000 | 0.0884          | 52.3566 |
 
 
 ### Framework versions

generation_config.json

@@ -1,4 +1,46 @@
 {
+  "alignment_heads": [
+    [
+      5,
+      3
+    ],
+    [
+      5,
+      9
+    ],
+    [
+      8,
+      0
+    ],
+    [
+      8,
+      4
+    ],
+    [
+      8,
+      7
+    ],
+    [
+      8,
+      8
+    ],
+    [
+      9,
+      0
+    ],
+    [
+      9,
+      7
+    ],
+    [
+      9,
+      9
+    ],
+    [
+      10,
+      5
+    ]
+  ],
   "begin_suppress_tokens": [
     220,
     50257
@@ -121,7 +163,7 @@
   "max_initial_timestamp_index": 1,
   "max_length": 448,
   "no_timestamps_token_id": 50363,
-  "pad_token_id": 50256,
+  "pad_token_id": 50257,
   "return_timestamps": false,
   "suppress_tokens": [
     1,
@@ -207,6 +249,8 @@
     49870,
     50254,
     50258,
+    50358,
+    50359,
     50360,
     50361,
     50362

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8f28ffa9ff48a79fc405b0edce95d1bec5dd84405c5f1da72ff15eaa3aa59eab
+oid sha256:680cb4b6859b3bbd3a69de3f04bac349f8df3060686132bef260e1ff2ddb3413
 size 967102729