AIbase
EN
Home
Whisper Small Ft Common Language Id
W

Whisper Small Ft Common Language Id

Developed by sanchit-gandhi
A general language identification model fine-tuned based on openai/whisper-small, achieving 88.6% accuracy on the evaluation dataset
Audio Classification
Transformers
Open Source License:Apache-2.0#Speech Language Recognition#Multilingual Classification#High Accuracy
Downloads 256.20k
Release Time : 2/24/2023

Model Overview

This model is an audio classification model based on the Whisper-small architecture, specifically designed for general language identification tasks. By fine-tuning on a general language dataset, it achieves high accuracy in language identification.

Model Features

High Accuracy
Achieves 88.6% identification accuracy on the evaluation dataset
Based on Whisper Architecture
Utilizes the proven Whisper-small architecture for fine-tuning
Multilingual Support
Suitable for general language identification (specific languages not explicitly stated)

Model Capabilities

Audio Classification
Language Identification
Speech Processing

Use Cases

Speech Processing
Multilingual Speech Recognition Preprocessing
Identify the language type of speech before the speech recognition process
Can improve the accuracy of subsequent speech recognition
Language Classification Service
Build a service to automatically identify the language type of speech
Accuracy reaches 88.6%
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