metadata

license: apache-2.0
tags:
  - generated_from_trainer
datasets:
  - common_voice_11_0
metrics:
  - wer
base_model: juancopi81/whisper-medium-es-train-valid
model-index:
  - name: juancopi81/whisper-medium-es-train-valid
    results:
      - task:
          type: automatic-speech-recognition
          name: Automatic Speech Recognition
        dataset:
          name: Common Voice 11.0
          type: mozilla-foundation/common_voice_11_0
          config: es
          split: test
          args: es
        metrics:
          - type: wer
            value: 6.15482563276337
            name: Wer

juancopi81/whisper-medium-es-train-valid

This model is a fine-tuned version of juancopi81/whisper-medium-es-train-valid on the common_voice_11_0 dataset. It achieves the following results on the evaluation set:

Loss: 0.2227
Wer: 6.1548

Using the script provided in the Whisper Sprint (Dec. 2022) the models achieves these results on the evaluation sets (WER):

google/fleurs: 6.94
mozilla-foundation/common_voice_11_0: XXXX

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: 32
eval_batch_size: 16
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 500
training_steps: 5000
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.0539	1.01	1000	0.2100	6.4465
0.0211	2.01	2000	0.2286	6.5082
0.0088	3.02	3000	0.2418	6.3848
0.0205	4.02	4000	0.2288	6.6603
0.1031	5.03	5000	0.2227	6.1548

Framework versions

Transformers 4.26.0.dev0
Pytorch 1.13.0+cu117
Datasets 2.7.1.dev0
Tokenizers 0.13.2