AIbase
EN
Home
Whisper Medium Jp
W

Whisper Medium Jp

Developed by vumichien
Japanese speech recognition model fine-tuned on the common_voice_11_0 dataset based on openai/whisper-medium
Speech Recognition
Transformers
 JapaneseOpen Source License:Apache-2.0#Japanese Speech Recognition#Low Word Error Rate#Multi-scenario Adaptation
Downloads 4,542
Release Time : 12/7/2022

Model Overview

This is an optimized automatic speech recognition (ASR) model for Japanese, fine-tuned on the Common Voice 11.0 Japanese dataset, capable of converting Japanese speech into text.

Model Features

Japanese Optimization
Specially fine-tuned for Japanese speech recognition, with excellent performance on Japanese test sets
Low Word Error Rate
Achieves a word error rate (WER) of only 9.04% on the Common Voice Japanese test set
Multi-dataset Validation
Performance evaluated on both Common Voice and Fleurs Japanese test sets

Model Capabilities

Japanese Speech Recognition
Speech-to-Text
Automatic Speech Transcription

Use Cases

Speech Transcription
Japanese Meeting Minutes
Automatically convert Japanese meeting recordings into text transcripts
Approximately 90% accuracy
Japanese Podcast Transcription
Transcribe Japanese podcast content into text
Voice Assistants
Japanese Voice Command Recognition
Used for command recognition systems in Japanese voice assistants
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
EnglishįŽ€äŊ“中文įšéĢ”中文ãĢãģんご
Š 2025AIbase