---
language:
- multilingual
license: apache-2.0
base_model: openai/whisper-small
tags:
- hf-asr-leaderboard
- generated_from_trainer
datasets:
- mozilla-foundation/common_voice_13_0
model-index:
- name: Whisper Small 3Lang 1 token only
  results: []
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/taiwanlingyj/cascasc/runs/zny1tow7)
# Whisper Small 3Lang 1 token only

This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the Common Voice 13.0 dataset.
It achieves the following results on the evaluation set:
- eval_loss: 0.7394
- eval_cer: 21.6707
- eval_wer: 34.9815
- eval_both_er: 29.2571
- eval_LID: 100.0
- eval_total: 29.2571
- eval_runtime: 5490.853
- eval_samples_per_second: 17.277
- eval_steps_per_second: 1.08
- epoch: 6.0794
- step: 162000

## 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: 4
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- training_steps: 240000
- mixed_precision_training: Native AMP

### Framework versions

- Transformers 4.42.4
- Pytorch 2.3.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1