AIbase
EN
Home
Wav2vec2 Xlsr 53 Rm Vallader With Lm
W

Wav2vec2 Xlsr 53 Rm Vallader With Lm

Developed by anuragshas
Romansh Vallader dialect speech recognition model based on wav2vec2-xlsr-53 with language model support
Speech Recognition
Transformers
Open Source License:Apache-2.0#Romansh speech recognition#Multi-dialect support#Low word error rate
Downloads 16
Release Time : 3/2/2022

Model Overview

This model is a speech recognition model based on the wav2vec2-xlsr-53 architecture, specifically fine-tuned for the Romansh Vallader dialect and integrated with a language model to improve recognition accuracy.

Model Features

Multilingual pre-training foundation
Based on the wav2vec2-xlsr-53 architecture, which has been pre-trained on 53 languages
Dialect-specific optimization
Specifically fine-tuned for the Romansh Vallader dialect
Language model integration
Integrated with a language model to improve recognition accuracy
Relatively high recognition accuracy
Achieved a word error rate of 32.06% on the evaluation set

Model Capabilities

Speech recognition
Dialect speech processing
Audio-to-text

Use Cases

Speech transcription
Vallader dialect recording transcription
Convert Romansh Vallader dialect speech recordings to text
Word error rate 32.06%
Language research
Dialect speech analysis
Used in linguistic research to analyze phonetic features of the Vallader dialect
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