Vamsi1801/whisper-small-te-updated

This is a fine-tuned model is using the OpenAI/whisper-small - 242M parameters.

The fine-tuned model uses a native Telugu dataset with a set of different audio files with various dialects/accents of the telugu language.

Being a small version, the model takes 1 hour and 15 minutes to train.

Steps to download this model locally on your respective system:

1. Using the "hugging face cli download" you can install the model locally which is directly from the hub.
2. The below code snippet can be directly run on the terminal:

• huggingface-cli download Vamsi1801/whisper-small-te-updated

3. The model is of size : 967M, when installed locally.