W

Wav2vec2 Base Japanese Asr

Developed by TKU410410103
A speech recognition model fine-tuned on the common_voice_11_0 Japanese dataset based on rinna/japanese-wav2vec2-base, supporting only hiragana output
Downloads 68
Release Time : 4/14/2024

Model Overview

This is a Japanese speech recognition model specifically optimized for Japanese speech, capable of converting Japanese speech into hiragana text.

Model Features

Japanese-specific speech recognition
A recognition model specifically optimized for Japanese speech
Hiragana output
The model outputs in hiragana format, suitable for Japanese text processing
Based on wav2vec2 architecture
Trained using the efficient wav2vec2-base architecture

Model Capabilities

Japanese speech recognition
Speech-to-text
Hiragana conversion

Use Cases

Speech transcription
Japanese speech to text
Convert Japanese speech content into hiragana text
WER of 14.18% on the common_voice_11_0 test set
Voice assistants
Japanese voice command recognition
Recognize Japanese voice commands and convert them to text
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
© 2025AIbase