README.md

metadata

license: apache-2.0
base_model: facebook/wav2vec2-base
tags:
  - automatic-speech-recognition
  - timit_asr
  - generated_from_trainer
datasets:
  - timit_asr
metrics:
  - wer
model-index:
  - name: wav2vec2-base-timit-fine-tuned
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: TIMIT_ASR - NA
          type: timit_asr
          config: clean
          split: test
          args: 'Config: na, Training split: train, Eval split: test'
        metrics:
          - name: Wer
            type: wer
            value: 0.4090867704634435

wav2vec2-base-timit-fine-tuned

This model is a fine-tuned version of facebook/wav2vec2-base on the TIMIT_ASR - NA dataset. It achieves the following results on the evaluation set:

• Loss: 0.4218
• Wer: 0.4091

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: 0.0001
• train_batch_size: 32
• eval_batch_size: 1
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 1000
• num_epochs: 20.0
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
3.1612	0.8621	100	3.1181	1.0
2.978	1.7241	200	2.9722	1.0
2.9185	2.5862	300	2.9098	1.0
2.1282	3.4483	400	2.0066	1.0247
1.1234	4.3103	500	1.0197	0.8393
0.602	5.1724	600	0.6714	0.6600
0.5032	6.0345	700	0.5285	0.5659
0.3101	6.8966	800	0.4819	0.5282
0.3432	7.7586	900	0.4653	0.5272
0.1922	8.6207	1000	0.4672	0.4918
0.2284	9.4828	1100	0.4834	0.4870
0.1372	10.3448	1200	0.4380	0.4727
0.1105	11.2069	1300	0.4509	0.4594
0.0992	12.0690	1400	0.4196	0.4544
0.1226	12.9310	1500	0.4237	0.4321
0.1013	13.7931	1600	0.4113	0.4298
0.0661	14.6552	1700	0.4038	0.4276
0.0901	15.5172	1800	0.4321	0.4225
0.053	16.3793	1900	0.4076	0.4236
0.0805	17.2414	2000	0.4336	0.4156
0.049	18.1034	2100	0.4193	0.4114
0.0717	18.9655	2200	0.4139	0.4091
0.0389	19.8276	2300	0.4216	0.4087

Framework versions

• Transformers 4.42.0.dev0
• Pytorch 2.3.0a0+git71dd2de
• Datasets 2.19.1
• Tokenizers 0.19.1