W

Whisper Large V2 French

Developed by bofenghuang
French speech recognition model fine-tuned from openai/whisper-large-v2, trained on over 2200 hours of French audio data
Downloads 103
Release Time : 1/11/2023

Model Overview

This model is optimized for French Automatic Speech Recognition (ASR) tasks, demonstrating excellent performance across multiple French speech datasets without predicting capitalization or punctuation.

Model Features

Multi-dataset training
Incorporates multiple high-quality French speech datasets including Common Voice 11.0, Multilingual LibriSpeech, and Voxpopuli
High performance
Significantly lower Word Error Rate (WER) than the base model across multiple test sets
Broad applicability
Supports recognition of both standard French and African-accented French

Model Capabilities

French speech-to-text conversion
High-accuracy speech recognition
Handling various French accents

Use Cases

Speech transcription
French meeting minutes
Convert French meeting recordings into text transcripts
Word error rate below 9%
French media content subtitling
Automatically generate subtitles for French videos
Approximately 5% word error rate on standard French content
Voice assistants
French voice command recognition
Used for voice command recognition in French voice assistants or smart home systems
Performs well with various accents
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
Š 2025AIbase