https://github.com/daisyyedda/whisper-large-v2-atcosim_corpus

A fine-tuned Whisper model (whisper-large-v2) for aviation audio transcription. WER < 5%.
https://github.com/daisyyedda/whisper-large-v2-atcosim_corpus

asr-model nlp whisper whisper-ai

Last synced: 3 months ago
JSON representation

A fine-tuned Whisper model (whisper-large-v2) for aviation audio transcription. WER < 5%.

• Host: GitHub
• URL: https://github.com/daisyyedda/whisper-large-v2-atcosim_corpus
• Owner: daisyyedda
• Created: 2024-07-10T18:12:40.000Z (6 months ago)
• Default Branch: main
• Last Pushed: 2024-07-15T20:26:04.000Z (6 months ago)
• Last Synced: 2024-09-27T06:22:13.736Z (4 months ago)
• Topics: asr-model, nlp, whisper, whisper-ai
• Language: Jupyter Notebook
• Homepage: https://huggingface.co/daisyyedda/whisper-large-v2-atcosim_corpus/
• Size: 90.8 KB
• Stars: 2
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# whisper-large-v2-atcosim_corpus

This model is a fine-tuned version of [openai/whisper-large-v2](https://huggingface.co/openai/whisper-large-v2) on the [atcosim_corpus](https://huggingface.co/datasets/Jzuluaga/atcosim_corpus) dataset.

## Training hyperparameters

The following hyperparameters were used during training:

- learning_rate: 1e-05

- train_batch_size: 32

- eval_batch_size: 8

- seed: 42

- distributed_type: multi-GPU

- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08

- lr_scheduler_type: linear

- lr_scheduler_warmup_steps: 250

- num_epochs: 50

  

## Evaluation results

- Loss: 0.1606

- Wer: 4.6858

## Framework versions

- Transformers 4.42.3

- Pytorch 2.3.1+cu118

- Datasets 2.20.0

- Tokenizers 0.19.1