Whisper Sinhala (Fine-Tuned)
This model is a fine-tuned version of openai/whisper-base on a Sinhala-accented English speech dataset.
🧾 Dataset
- Source: Sinhala Accent Voice Data
- Language: English (Sri Lankan / Sinhala accent)
- Audio Format: WAV, mono, 16kHz
- Split: train/dev sets
🎯 Objective
The model was trained for Automatic Speech Recognition (ASR) on Sri Lankan-accented English, using Whisper’s decoder-based architecture.
💡 Intended Use
- Voice-to-text applications for Sri Lankan English speakers
- Improving ASR robustness in regional accents
- Research on accent adaptation in low-resource settings
🏷️ Tags
sinhala whisper asr accent english finetuned sri-lanka
- Downloads last month
- 1
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support