N

Nb Whisper Large

Developed by NbAiLabBeta
An automatic speech recognition model developed by the National Library of Norway, based on the Whisper architecture, supporting speech transcription and translation of Norwegian and English.
Downloads 776
Release Time : 1/9/2024

Model Overview

NB-Whisper Large is a cutting - edge automatic speech recognition (ASR) and speech translation model developed based on OpenAI Whisper. It has been trained on 66,000 hours of Norwegian data and is suitable for high - precision speech - to - text tasks.

Model Features

Multi - size model series
Provides five models with different parameter scales from Tiny (39M) to Large (1550M) to meet the needs of different computing resources
Professional variant versions
Provides two professional variants, verbatim and semantic, suitable for precise transcription and content summarization scenarios respectively
Norwegian optimization
Trained on 66,000 hours of Norwegian data, with special optimization for Norwegian dialects and accents
Multi - format support
Provides PyTorch, TensorFlow, ONNX, and ggml format of whisper.cpp, supporting multiple deployment methods

Model Capabilities

Norwegian speech transcription
English speech transcription
Speech translation
Speaker separation
Timestamp alignment

Use Cases

Media processing
Broadcast content transcription
Automatically convert the broadcast programs of the Norwegian Broadcasting Corporation (NRK) into manuscripts
Supports generating subtitle files with timestamps
Meeting minutes
Automatically record the speech content of the Norwegian Parliament
The semantic version can generate a concise meeting summary
Education and research
Linguistic research
Analyze the speech characteristics of Norwegian dialects
The verbatim version provides precise phoneme - level transcription
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
© 2025AIbase