W

Wav2vec2 Xlsr Dhivehi

Developed by sammy786
This is an automatic speech recognition (ASR) model fine-tuned on Dhivehi speech datasets based on the facebook/wav2vec2-xls-r-1b model.
Downloads 30
Release Time : 3/2/2022

Model Overview

This model is specifically optimized for Dhivehi and is used for automatic speech recognition tasks that convert speech to text.

Model Features

Dhivehi optimization
Specifically fine-tuned for Dhivehi, providing better speech recognition performance
Based on a large model
Fine-tuned on the facebook/wav2vec2-xls-r-1b large model, inheriting its powerful speech feature extraction capabilities
High performance
Achieved 26.91% WER and 4.02% CER on the Common Voice 8 test set

Model Capabilities

Dhivehi speech recognition
Speech-to-text
Supports long speech processing

Use Cases

Speech transcription
Dhivehi speech transcription
Convert Dhivehi speech content into text
Accuracy rate of 73.09% (1-WER)
Voice assistant
Dhivehi voice interaction
Used to build Dhivehi voice assistants
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
Š 2025AIbase