W

Wav2vec2 Large Xlsr 53 Th

Developed by airesearch
This is an automatic speech recognition (ASR) model fine-tuned on the Common Voice 7.0 Thai dataset based on the wav2vec2-large-xlsr-53 model.
Downloads 110.74k
Release Time : 3/2/2022

Model Overview

This model is specifically designed for Thai speech recognition tasks, fine-tuned on the Common Voice 7.0 Thai dataset, and supports multiple Thai tokenizers.

Model Features

Multi-tokenizer Support
Integrates various Thai tokenizers such as PyThaiNLP and deepcut to improve recognition accuracy
High Performance
Achieved low WER (0.9524%) and CER (0.1623%) on the Common Voice 7.0 test set
Data Cleaning Optimization
Uses specially designed cleaning rules to preprocess the dataset, enhancing model training effectiveness

Model Capabilities

Thai Speech Recognition
Speech-to-Text
Supports multiple Thai tokenization methods

Use Cases

Speech Transcription
Thai Speech to Text
Convert Thai speech content into text format
Achieved a WER of 0.9524% on the test set
Voice Assistants
Thai Voice Command Recognition
Used for Thai voice assistants or smart device command recognition
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
© 2025AIbase