Fine-tuning API/UI to support Whisper

[programmaticsoftware]

Hi,
Is there any plan for the Fine-tuning API/UI to support Finetuning Whisper?

[gongouveia]

@programmaticsoftware Hello, you can use my UI WhisperTemple to generate new audio samples, and edit the transcriptions. Also it is created to easily train/ fine tune on the new samples.
Please check it out, support it and request new features.

[programmaticsoftware]

Hi @gongouveia
Can I use this UI if I'm not familiar with Python?
Can I use it from a REST API?
Thank you

[gongouveia]

@gongouveia you just need to download the files, and run the UI from the console and for the training you can also use the console.

Please, if you can describe better what are your needs, I can direct you better for diferent solutions.

[programmaticsoftware]

@gongouveia

I have a dataset of about 500 audio-text pair samples to fine-tune whisper, and I'm looking If I need to hire a Python ML expert to do the fine-tuning or I can use a REST API or a UI.

Thank you

[gongouveia]

@programmaticsoftware there is no UI or API that I am aware of to train the Whisper. If you need someone to finentune it on your data shoot me a message, I would be glad to help.

[programmaticsoftware]

@gongouveia I've sent you an email using the email address I found on your website.