AIbase
EN
Home
Whisper Medium Ko Zeroth
W

Whisper Medium Ko Zeroth

Developed by seastar105
A speech recognition model fine-tuned on the Zeroth Korean dataset based on OpenAI Whisper Medium model, with a word error rate of 3.64%
Speech Recognition
Transformers
 KoreanOpen Source License:Apache-2.0#Korean speech recognition#Low word error rate#Whisper architecture
Downloads 154
Release Time : 12/11/2022

Model Overview

This is an automatic speech recognition (ASR) model optimized for Korean, fine-tuned based on the Whisper Medium architecture, suitable for Korean speech-to-text tasks

Model Features

Low word error rate
Achieves a word error rate of 3.64% on the Zeroth Korean test set
Korean optimization
Specially fine-tuned for Korean speech data to improve recognition accuracy
Based on Whisper architecture
Utilizes the advanced speech recognition architecture of OpenAI Whisper

Model Capabilities

Korean speech recognition
Speech-to-text
Automatic speech recognition

Use Cases

Speech transcription
Korean meeting minutes
Automatically converts Korean meeting recordings into text transcripts
Highly accurate transcript text
Korean subtitle generation
Automatically generates subtitles for Korean video content
Accurate subtitles with 3.64% word error rate
Voice assistants
Korean voice command recognition
Used for understanding voice commands in Korean voice assistant systems
High accuracy command recognition
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