The Dogri ASR (Automatic Speech Recognition) model is designed to convert spoken Dogri into text from 16kHz mono-channel audio files. It is built on a Conformer-Large architecture with 120 million parameters, providing high accuracy in speech recognition. The model utilizes a hybrid CTC-RNNT decoder to improve transcription quality. With 17 conformer blocks and a model dimension of 512, it effectively captures linguistic and acoustic features, making it ideal for speech-to-text applications in Dogri.
MIT
Sector Agnostic
AI4Bharat
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
Open
21/02/25 13:21:47
N.A.
0
© 2026 - Copyright AIKosh. All rights reserved.