End of training

README.md

@@ -1,7 +1,7 @@
 ---
 language:
 - hi
-base_model: nurzhanit/dlc_audio
+base_model: nurzhanit/whisper-enhanced-ml
 tags:
 - hf-asr-leaderboard
 - generated_from_trainer
@@ -18,6 +18,8 @@ model-index:
     dataset:
       name: Common Voice 11.0
       type: mozilla-foundation/common_voice_11_0
+      config: default
+      split: None
       args: 'config: hi, split: test'
     metrics:
     - name: Wer
@@ -30,9 +32,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # Whisper Small Hi - Sanchit Gandhi
 
-This model is a fine-tuned version of [nurzhanit/dlc_audio](https://huggingface.co/nurzhanit/dlc_audio) on the Common Voice 11.0 dataset.
+This model is a fine-tuned version of [nurzhanit/whisper-enhanced-ml](https://huggingface.co/nurzhanit/whisper-enhanced-ml) on the Common Voice 11.0 dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0000
+- Loss: 0.0001
 - Wer: 0.0
 
 ## Model description
@@ -58,28 +60,38 @@ The following hyperparameters were used during training:
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- lr_scheduler_warmup_steps: 20
-- training_steps: 500
+- lr_scheduler_warmup_steps: 100
+- training_steps: 1000
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss | Wer |
-|:-------------:|:-----:|:----:|:---------------:|:---:|
-| 0.0002        | 50.0  | 50   | 0.0001          | 0.0 |
-| 0.0001        | 100.0 | 100  | 0.0001          | 0.0 |
-| 0.0           | 150.0 | 150  | 0.0000          | 0.0 |
-| 0.0           | 200.0 | 200  | 0.0000          | 0.0 |
-| 0.0           | 250.0 | 250  | 0.0000          | 0.0 |
-| 0.0           | 300.0 | 300  | 0.0000          | 0.0 |
-| 0.0           | 350.0 | 350  | 0.0000          | 0.0 |
-| 0.0           | 400.0 | 400  | 0.0000          | 0.0 |
-| 0.0           | 450.0 | 450  | 0.0000          | 0.0 |
-| 0.0           | 500.0 | 500  | 0.0000          | 0.0 |
+| Training Loss | Epoch    | Step | Validation Loss | Wer    |
+|:-------------:|:--------:|:----:|:---------------:|:------:|
+| 0.2307        | 8.3333   | 50   | 0.0742          | 9.7521 |
+| 0.0048        | 16.6667  | 100  | 0.0022          | 2.1488 |
+| 0.0005        | 25.0     | 150  | 0.0004          | 0.0    |
+| 0.0003        | 33.3333  | 200  | 0.0003          | 0.0    |
+| 0.0002        | 41.6667  | 250  | 0.0002          | 0.0    |
+| 0.0002        | 50.0     | 300  | 0.0002          | 0.0    |
+| 0.0002        | 58.3333  | 350  | 0.0001          | 0.0    |
+| 0.0001        | 66.6667  | 400  | 0.0001          | 0.0    |
+| 0.0001        | 75.0     | 450  | 0.0001          | 0.0    |
+| 0.0001        | 83.3333  | 500  | 0.0001          | 0.0    |
+| 0.0001        | 91.6667  | 550  | 0.0001          | 0.0    |
+| 0.0001        | 100.0    | 600  | 0.0001          | 0.0    |
+| 0.0001        | 108.3333 | 650  | 0.0001          | 0.0    |
+| 0.0001        | 116.6667 | 700  | 0.0001          | 0.0    |
+| 0.0001        | 125.0    | 750  | 0.0001          | 0.0    |
+| 0.0001        | 133.3333 | 800  | 0.0001          | 0.0    |
+| 0.0001        | 141.6667 | 850  | 0.0001          | 0.0    |
+| 0.0001        | 150.0    | 900  | 0.0001          | 0.0    |
+| 0.0001        | 158.3333 | 950  | 0.0001          | 0.0    |
+| 0.0001        | 166.6667 | 1000 | 0.0001          | 0.0    |
 
 
 ### Framework versions
 
-- Transformers 4.35.2
-- Pytorch 2.1.1+cu121
-- Datasets 2.15.0
-- Tokenizers 0.15.0
+- Transformers 4.40.0
+- Pytorch 2.5.0+cu124
+- Datasets 3.0.2
+- Tokenizers 0.19.1

generation_config.json

@@ -10,5 +10,5 @@
   "pad_token_id": 50257,
   "return_timestamps": false,
   "suppress_tokens": [],
-  "transformers_version": "4.35.2"
+  "transformers_version": "4.40.0"
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1481e1c08d8705d54add94920caa14481ff573bc2f38bf871bde249af3fd212d
+oid sha256:ef39c9dc185c716656f9d4124c84283018897c9a7dc25ec44adc6ec2dc9b4a8f
 size 966995080