Logo Lanfrica

Brian-Mwangi-developer/whisper-finetune-swahili-commonvoice

Domain:

natural language processing

Record type:

software
Creator:
Bri
Host:
Finetuning whisper small model on common voice swahili data ## Whisper Fine-Tuning on Common Voice Swahili Dataset This repository contains the code and steps to fine-tune OpenAI's Whisper model on the Common Voice Swahili dataset using the Hugging Face transformers library. ## Model Overview Whisper is a general-purpose speech recognition model developed by OpenAI. It is capable of transcribing multilingual audio and fine-tuning it on specific datasets for improved performance in targeted languages. In this project, we fine-tune Whisper on the Swahili subset of the Common Voice dataset.

Languages