AIbase
EN
Home
Wav2vec2 Random
W

Wav2vec2 Random

Developed by patrickvonplaten
An automatic speech recognition model fine-tuned on the TIMIT_ASR dataset based on the wav2vec2-base-random model
Speech Recognition
Transformers
#English speech recognition#TIMIT dataset fine-tuning#Low-resource optimization
Downloads 16
Release Time : 3/2/2022

Model Overview

This model is an implementation of the wav2vec2 architecture for English speech recognition, fine-tuned on the TIMIT_ASR dataset, capable of converting speech to text

Model Features

Based on wav2vec2 architecture
Utilizes the self-supervised learning architecture of wav2vec2 proposed by Facebook Research
Fine-tuned on TIMIT_ASR dataset
Fine-tuned on the standard TIMIT speech recognition dataset
Medium-sized model
Based on the wav2vec2-base architecture, suitable for environments with moderate computational resources

Model Capabilities

English speech recognition
Speech-to-text conversion

Use Cases

Speech transcription
Speech recording transcription
Convert English speech recordings into text transcripts
Achieves a word error rate of 0.8364 on the TIMIT evaluation set
Voice interface
Voice command recognition
Recognize simple English voice commands
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