AIbase
EN
Home
Wav2vec2 Xlsr 1B NPSC NN
W

Wav2vec2 Xlsr 1B NPSC NN

Developed by NbAiLab
An automatic speech recognition model fine-tuned on the Nynorsk (NN) speech dataset based on facebook/wav2vec2-xls-r-1b
Speech Recognition
Transformers
Open Source License:Apache-2.0#Nynorsk speech recognition#Low word error rate#Large model fine-tuning
Downloads 22
Release Time : 3/2/2022

Model Overview

This is a speech recognition model optimized for Nynorsk, fine-tuned on the NPSC dataset with low word error rate and character error rate.

Model Features

Low error rate
Achieves 13.35% word error rate and 4.54% character error rate on the Nynorsk test set
Large model foundation
Fine-tuned based on facebook's wav2vec2-xls-r-1b large model with powerful speech feature extraction capabilities
Professional dataset training
Trained using the NbAiLab/NPSC professional speech dataset, optimized for Nynorsk

Model Capabilities

Nynorsk speech recognition
16KHz audio processing
High-precision transcription

Use Cases

Speech transcription
Meeting minutes
Convert Nynorsk meeting recordings into text transcripts
High-accuracy transcription results
Media subtitle generation
Automatically generate subtitles for Nynorsk video content
Reduces manual subtitle production workload
Speech analysis
Speech data analysis
Analyze linguistic features in Nynorsk speech data
Supports linguistic research and speech analysis
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