W

Wav2vec2 Xls R 300m Pa IN R5

Developed by DrishtiSharma
This is an automatic speech recognition model fine-tuned on the Punjabi (India) dataset based on the facebook/wav2vec2-xls-r-300m model.
Downloads 25
Release Time : 3/2/2022

Model Overview

This model is specifically designed for Punjabi (India) speech recognition tasks, trained on the Mozilla Common Voice 8.0 dataset, and can be used to convert Punjabi speech into text.

Model Features

Punjabi speech recognition
Speech recognition model specifically optimized for Punjabi (India)
Based on wav2vec2 architecture
Uses facebook's wav2vec2-xls-r-300m pre-trained model as the foundation
Trained on Common Voice dataset
Fine-tuned using Mozilla Foundation's common_voice_8_0 dataset

Model Capabilities

Punjabi speech-to-text
Automatic speech recognition

Use Cases

Speech transcription
Punjabi speech transcription
Convert Punjabi speech content into text
Achieved WER of 41.87% and CER of 13.30% on the test set
Voice assistant
Punjabi voice assistant
Provides voice interaction capabilities for Punjabi users
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
Š 2025AIbase