W

Wav2vec2 Xls R 300m Cv8 Da

Developed by saattrupdan
A speech recognition model fine-tuned on Common Voice 8.0 Danish data based on facebook/wav2vec2-xls-r-300m
Downloads 13
Release Time : 3/2/2022

Model Overview

This model is an automatic speech recognition (ASR) model for Danish, fine-tuned from a multilingual pre-trained model, suitable for Danish speech-to-text tasks.

Model Features

Multilingual pre-training foundation
Fine-tuned from the powerful multilingual wav2vec2-xls-r-300m model, with excellent speech feature extraction capabilities
Danish optimization
Specifically fine-tuned for Danish, performing well on Danish speech recognition tasks
Lightweight model
With 300M parameters, the model size is relatively moderate, suitable for practical deployment

Model Capabilities

Danish speech recognition
Speech-to-text

Use Cases

Speech transcription
Danish speech transcription
Convert Danish speech content into text
WER of 26.45 on the Common Voice 8.0 test set
Voice assistants
Danish voice command recognition
Used for voice command recognition in Danish voice assistants or control systems
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
Š 2025AIbase