Downloads · 30 days
0
pevers-whisperd-nl-whispercpp-q8_0-no-speaker-tags
terriblekiwi/pevers-whisperd-nl-whispercpp-q8_0-no-speaker-tags
pevers-whisperd-nl-whispercpp-q8_0-no-speaker-tags is a automatic speech recognition model from terriblekiwi. Use it when you need speech turned into text. The card lists the license as mit.
This repository contains a whisper.cpp-compatible ggml q80 derivative of:
Access
Public
Updated May 23, 2026
Repo size
1.7 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.bin1.7 GB · 100%
At a glance
- Task
- Automatic Speech Recognition
- License
- mit
- Access
- Public
- Created
- May 22, 2026
- Updated
- May 23, 2026
- SHA
- 9fb62502
Base models
Model card
From the Hugging Face model README
pevers whisperd nl whisper.cpp q8_0 no-speaker-tags
This repository contains a whisper.cpp-compatible ggml q8_0 derivative of:
- upstream fine-tune: pevers/whisperd-nl
- base model: openai/whisper-large-v3
File
ggml-pevers-whisperd-nl-whispercpp-q8_0-no-speaker-tags.bin
Important limitation
This published variant is the working stock-whisper.cpp Android compatibility build.
It does not preserve the original upstream speaker/event tag behavior. The extra speaker and event tokens from the upstream tokenizer were removed in this variant so it can run in the current runtime without crashing.
If you need full original speaker/event tags, use the upstream model or a future full-tag-compatible whisper.cpp build instead of this artifact.
Attribution
- Original fine-tune: pevers/whisperd-nl
- Base model family: openai/whisper-large-v3
- Runtime/conversion ecosystem: ggml-org/whisper.cpp
- Task
- Automatic Speech Recognition
- License
- mit
- Languages
- nl
- Created
- May 22, 2026
- Updated
- May 23, 2026