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: 0.18583042973286876

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: 0.1858

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: 50
training_steps: 1000

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.0142	0.8621	50	0.0076	1.3937
0.0086	1.7241	100	0.0054	1.1614
0.0067	2.5862	150	0.0032	0.5575
0.0032	3.4483	200	0.0019	1.0453
0.0018	4.3103	250	0.0011	0.3252
0.0016	5.1724	300	0.0012	1.3240
0.0024	6.0345	350	0.0007	0.2787
0.0014	6.8966	400	0.0004	0.1858
0.0012	7.7586	450	0.0002	0.1858
0.0003	8.6207	500	0.0003	0.1858
0.0001	9.4828	550	0.0002	0.1858
0.0004	10.3448	600	0.0002	0.1858
0.0001	11.2069	650	0.0002	0.1858
0.0004	12.0690	700	0.0002	0.1858
0.0004	12.9310	750	0.0002	0.1858
0.0002	13.7931	800	0.0002	0.1858
0.0003	14.6552	850	0.0002	0.1858
0.0001	15.5172	900	0.0001	0.1858
0.0001	16.3793	950	0.0001	0.1858
0.0001	17.2414	1000	0.0001	0.1858

Framework versions

Transformers 4.40.0
Pytorch 2.5.0+cu124
Datasets 3.0.2
Tokenizers 0.19.1