metadata

language:
  - hi
base_model: nurzhanit/whisper-enhanced-ml
tags:
  - hf-asr-leaderboard
  - generated_from_trainer
datasets:
  - mozilla-foundation/common_voice_11_0
metrics:
  - wer
model-index:
  - name: Whisper Small Hi - Sanchit Gandhi
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        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
            type: wer
            value: 35.622895622895626

Whisper Small Hi - Sanchit Gandhi

This model is a fine-tuned version of nurzhanit/whisper-enhanced-ml on the Common Voice 11.0 dataset. It achieves the following results on the evaluation set:

Loss: 0.0001
Wer: 35.6229

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 1e-05
train_batch_size: 16
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: 100
training_steps: 1000

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
3.2948	0.2688	50	2.0785	12.9293
1.4104	0.5376	100	1.1845	2.6263
0.5806	0.8065	150	0.3972	24.0404
0.0701	1.0753	200	0.0263	48.0471
0.0023	1.3441	250	0.0012	39.2593
0.0006	1.6129	300	0.0005	39.8653
0.0004	1.8817	350	0.0004	31.7508
0.0003	2.1505	400	0.0003	32.7609
0.0002	2.4194	450	0.0002	34.6801
0.0002	2.6882	500	0.0002	31.4141
0.0002	2.9570	550	0.0002	38.2155
0.0001	3.2258	600	0.0001	33.6364
0.0001	3.4946	650	0.0001	36.2290
0.0001	3.7634	700	0.0001	35.7239
0.0001	4.0323	750	0.0001	34.9158
0.0001	4.3011	800	0.0001	37.2727
0.0001	4.5699	850	0.0001	35.2862
0.0001	4.8387	900	0.0001	35.5892
0.0001	5.1075	950	0.0001	34.9158
0.0001	5.3763	1000	0.0001	35.6229

Framework versions

Transformers 4.40.0
Pytorch 2.5.0+cu124
Datasets 3.0.2
Tokenizers 0.19.1