X

Xls R 300m Pt

Developed by AlexN
This is an automatic speech recognition model fine-tuned on the Portuguese Common Voice 8.0 dataset based on Facebook's wav2vec2-xls-r-300m model
Downloads 28
Release Time : 3/2/2022

Model Overview

This model is specifically designed for Portuguese speech recognition tasks, performing excellently on the Common Voice 8.0 dataset with a word error rate (WER) of 19.36%

Model Features

Multilingual Support
Based on XLS-R architecture, supporting cross-language transfer learning
Efficient Performance
Achieves a word error rate of 19.36% on Portuguese test sets
Robustness
Stable performance on robust speech event test data

Model Capabilities

Portuguese speech-to-text
Automatic speech recognition
Handling speech with different accents

Use Cases

Speech Transcription
Portuguese Speech Transcription
Convert Portuguese speech to text
WER of 19.36% on Common Voice 8.0 test set
Voice Assistants
Portuguese Voice Command Recognition
Front-end speech recognition for Portuguese voice assistants
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
Š 2025AIbase