metadata

license: apache-2.0
tags:
  - generated_from_trainer
datasets:
  - PolyAI/minds14
metrics:
  - wer
model-index:
  - name: whisper-tiny-finetuned-minds14-en-v2
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: PolyAI/minds14
          type: PolyAI/minds14
          config: en-US
          split: train
          args: en-US
        metrics:
          - name: Wer
            type: wer
            value: 0.3695395513577332

whisper-tiny-finetuned-minds14-en-v2

This model is a fine-tuned version of openai/whisper-tiny on the PolyAI/minds14 dataset. It achieves the following results on the evaluation set:

Loss: 0.7064
Wer Ortho: 0.3720
Wer: 0.3695

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: 64
eval_batch_size: 16
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: constant_with_warmup
lr_scheduler_warmup_steps: 50
training_steps: 500

Training results

Training Loss	Epoch	Step	Validation Loss	Wer Ortho	Wer
1.5939	7.14	50	0.6729	0.4294	0.4032
0.1094	14.29	100	0.5254	0.3763	0.3642
0.0057	21.43	150	0.5993	0.3646	0.3607
0.002	28.57	200	0.6255	0.3609	0.3601
0.0013	35.71	250	0.6444	0.3652	0.3625
0.0009	42.86	300	0.6603	0.3689	0.3660
0.0007	50.0	350	0.6736	0.3701	0.3678
0.0006	57.14	400	0.6857	0.3726	0.3707
0.0005	64.29	450	0.6965	0.3708	0.3689
0.0004	71.43	500	0.7064	0.3720	0.3695

Framework versions

Transformers 4.30.2
Pytorch 2.0.1+cu117
Datasets 2.13.1
Tokenizers 0.13.3