X

Xls R 2B Te

Developed by chmanoj
This is a Telugu automatic speech recognition (ASR) model fine-tuned based on the facebook/wav2vec2-xls-r-2b model, trained on the OpenSLR SLR66 dataset
Downloads 20
Release Time : 3/2/2022

Model Overview

This model is specifically designed for Telugu speech recognition tasks, capable of converting Telugu speech into text

Model Features

High Accuracy Telugu Recognition
Achieves a word error rate (WER) of 0.51 and a character error rate (CER) of 0.097 on the test set
Based on XLS-R Architecture
Uses facebook's wav2vec2-xls-r-2b as the base model for fine-tuning
Trained on Public Dataset
Trained using the publicly available OpenSLR SLR66 dataset

Model Capabilities

Telugu speech recognition
Speech-to-text
Robust speech processing

Use Cases

Speech Transcription
Telugu Meeting Minutes
Automatically transcribe Telugu meeting recordings into text records
Accuracy approximately 90.3% (based on 1-CER)
Voice Assistant
Provide voice interaction functionality for Telugu-speaking users
Education
Language Learning Assistance
Help learners check the accuracy of Telugu pronunciation
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
Š 2025AIbase