W

Whisper Large Khmer Asr

Developed by ksoky
A Khmer automatic speech recognition model fine-tuned based on OpenAI Whisper Large, trained on the SLR42 dataset
Downloads 37
Release Time : 11/17/2022

Model Overview

This model is a Khmer fine-tuned version of Whisper Large, specifically designed for Khmer speech recognition tasks. It achieved a word error rate of 29.52% on the SLR42 test set.

Model Features

Khmer optimization
Specially fine-tuned for Khmer speech characteristics to improve recognition accuracy
Based on Whisper architecture
Utilizes OpenAI Whisper's powerful speech recognition capabilities as the base model
Open-source license
Released under Apache 2.0 license, allowing both commercial and research use

Model Capabilities

Khmer speech recognition
Speech-to-text conversion

Use Cases

Speech transcription
Khmer meeting minutes
Automatically transcribe Khmer meeting recordings into text records
Word error rate approximately 29.5%
Media subtitle generation
Automatically generate subtitles for Khmer video content
Voice assistants
Khmer voice command recognition
Used for command recognition in Khmer voice assistant systems
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
Š 2025AIbase