Getting Started

Get started with Speech-to-text-finetune Blueprint using one of the options below:

☁️ Google Colab (GPU)☁️ GitHub Codespaces💻 Local Installation

Finetune a STT model using CommonVoice data by launching the Google Colab notebook below

Click the button below to launch the project directly in Google Colab:

Click the button below to launch the project directly in GitHub Codespaces:

Once the Codespaces environment launches, inside the terminal, install dependencies:

pip install -e .

To load the app for making your own dataset for STT finetuning:

python src/speech_to_text_finetune/make_local_dataset_app.py

To load the Transcription app:

python demo/transcribe_app.py

To run the Finetuning job:

python src/speech_to_text_finetune/finetune_whisper.py

To install the project locally:

git clone https://github.com/mozilla-ai/speech-to-text-finetune.git
cd speech-to-text-finetune

install dependencies:

pip install -e .

To load the app for making your own dataset for STT finetuning:

python src/speech_to_text_finetune/make_local_dataset_app.py

To load the Transcription app:

python demo/transcribe_app.py

To run the Finetuning job:

python src/speech_to_text_finetune/finetune_whisper.py

Troubleshooting - TBA