dalavaihemanth/SWECHA_GONTHUKA_ASR_finetune

SWECHA GONTHUKA ASR Fine-tuned

This model is fine-tuned from swechatelangana/swecha-gonthuka-asr on a custom Telugu speech dataset.

Dataset

• 4911 Telugu audio recordings
• 16 kHz WAV files
• Multiple speakers

Base Model

swechatelangana/swecha-gonthuka-asr

Training

• Framework: Hugging Face Transformers
• Model: Wav2Vec2ForCTC

Usage

from transformers import Wav2Vec2Processor, Wav2Vec2ForCTC

processor = Wav2Vec2Processor.from_pretrained(
    "dalavaihemanth/SWECHA_GONTHUKA_ASR_finetune"
)

model = Wav2Vec2ForCTC.from_pretrained(
    "dalavaihemanth/SWECHA_GONTHUKA_ASR_finetune"
)