Whisper-Large-V2 Fine-Tuned on Jacob's Voice Data
This repository contains fine-tuned versions of the openai/whisper-large-v2 model, adapted specifically for Jacob's voice data from the Secret-Sleepover-Society. The fine-tuning process involved training the model on a dataset featuring Jacob's voice, enhancing its ability to accurately transcribe his speech.
Model Checkpoints
The following checkpoints are provided, each representing different stages of the fine-tuning process. The checkpoints are named based on the total number of training steps, with a batch size of 2:
- checkpoint-6717: After 1 epoch of fine-tuning
- checkpoint-10075: After 1.5 epochs of fine-tuning
- checkpoint-13433: After 2 epochs of fine-tuning
- checkpoint-16791: After 2.5 epochs of fine-tuning
- checkpoint-20149: After 3 epochs of fine-tuning
Each checkpoint reflects the model's state at the corresponding step, allowing users to choose the version that best suits their needs.