AIbase
EN
Home
Wav2vec2 Large Xlsr 53 Coraa Brazilian Portuguese Gain Normalization
W

Wav2vec2 Large Xlsr 53 Coraa Brazilian Portuguese Gain Normalization

Developed by alefiury
This is a Wav2vec 2.0 model fine-tuned for Portuguese, trained on multiple Portuguese speech datasets including CORAA, CETUC, MLS, etc.
Speech Recognition
Transformers
 OtherOpen Source License:Apache-2.0#Portuguese speech recognition#Multi-dataset training#Low word error rate
Downloads 28
Release Time : 3/27/2022

Model Overview

Based on the Wav2Vec 2.0 architecture, this model is specifically optimized for Portuguese speech recognition tasks, capable of converting Portuguese speech to text.

Model Features

Multi-dataset training
The model integrates multiple Portuguese datasets such as CORAA, CETUC, MLS, VoxForge, and Common Voice for training, improving recognition accuracy.
Low word error rate
Achieved a word error rate (WER) of 24.89% on the CORAA test set, demonstrating excellent performance.
XLSR architecture
Based on the large-scale cross-lingual speech representation learning (XLSR) Wav2Vec2 architecture, it has powerful speech feature extraction capabilities.

Model Capabilities

Portuguese speech recognition
Speech-to-text
Audio processing

Use Cases

Speech transcription
Automatic meeting transcription
Automatically convert Portuguese meeting recordings into text transcripts
24.89% WER
Voice assistant
Provide speech recognition capabilities for Portuguese voice assistants
Education
Language learning applications
Help learners practice Portuguese pronunciation and listening
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
English简体中文繁體中文にほんご
© 2025AIbase