Add NB-Whisper Norwegian transcription models#55
Open
androa wants to merge 1 commit into
Open
Conversation
Add NB-Whisper Small, Medium, and Large models to the transcription model catalog. These are Norwegian-tuned Whisper models by NB AI-Lab (National Library of Norway), trained on 66,000 hours of Norwegian speech data. Models use the standard GGML format and are downloaded from HuggingFace with pinned SHA256 checksums. They are always visible in settings (no feature flag). - NB-Whisper Small (~465 MB): fast with good Norwegian accuracy - NB-Whisper Medium (~1.4 GB): better accuracy, larger download - NB-Whisper Large (~2.9 GB): best accuracy for Norwegian speech Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Owner
|
Hi Androa. This is a bit too specific to merge in to the main, however I'm working on giving users more control on adding their own models for ASR. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Adds NB-Whisper models (Small, Medium, Large) to the transcription model catalog for improved Norwegian speech recognition.
NB-Whisper is developed by NB AI-Lab (National Library of Norway), fine-tuned on 66,000 hours of Norwegian speech data. The models use the standard GGML format, so no infrastructure changes are needed — the existing download, validation, and whisper.cpp transcription pipelines work as-is.
Models added
Changes
TranscriptionModels.swift: Added 3 NB-Whisper model entries with pinned HuggingFace download URLs, SHA256 checksums, and file sizes. Models are always visible in settings (no feature flag).TranscriptionModelCatalogTests.swift(new): Tests for catalog presence, unique IDs/filenames, metadata validity, and lookup.Usage
Select an NB-Whisper model in Settings → Transcription Model, and set the language to Norwegian for best results.