K

Kotoba Whisper V2.1

Developed by kotoba-tech
Kotoba-Whisper-v2.1 is a Japanese automatic speech recognition (ASR) model based on Whisper, integrating an additional post-processing stack that automatically adds punctuation marks.
Downloads 2,589
Release Time : 9/17/2024

Model Overview

This model focuses on Japanese speech recognition tasks, achieving automatic punctuation addition through the integration of the punctuators library, thereby enhancing the readability of transcribed text.

Model Features

Automatic Punctuation Addition
By integrating the punctuators library, the model can automatically add punctuation marks to transcribed text, improving readability.
Optimized Japanese Recognition
Specially optimized for Japanese speech recognition, it performs excellently on multiple Japanese datasets.
Pipeline Integration
The post-processing stack is seamlessly integrated through a pipeline, simplifying the usage process.

Model Capabilities

Japanese Speech Recognition
Automatic Punctuation Addition
Batch Audio Processing

Use Cases

Speech Transcription
Meeting Minutes Transcription
Convert Japanese meeting recordings into punctuated text transcripts
CER 17.7 (CommonVoice 8 Test Set)
Media Content Subtitle Generation
Automatically generate punctuated subtitles for Japanese video content
CER 15.4 (JSUT Basic 5000 Dataset)
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
Š 2025AIbase