license: apache-2.0
base_model: openai/whisper-base
tags:
- whisper
- whisper.cpp
- ggml
- automatic-speech-recognition
- windyword
- mobile
- quantized
library_name: whisper.cpp
Windy Lite — ggml q5_1 (mobile app shelf)
Promoted, not authored. This artifact was created in Windstorm Labs and pushed downstream unchanged. Do not edit weights here — fix them in the lab and re-promote, or this copy becomes a lie.
| File | ggml-windy-lite-q5_1.bin |
| Size | 59.7 MB |
| SHA-256 | 53aaa0467a2ecfe93692a870a41c52394bc3f17955245553fab66e3b05621ccf |
| Quantization | q5_1 |
Provenance — one hop back
| Promoted from | WindyWord/windy-lite-ggml |
| At revision | 61828f5e88760e69b43cd5d5b5bba42377a028a2 |
| Origin of record | WindstormLabs/windy-lite-ggml @ 12c051ef8a65e9e7f71b9c7234c498fafa85621a |
| Converted from | WindyWord/listen-windy-lite @ 11ef4da092108720ac3a3df147d58b326e899677 |
Full chain: WindstormLabs -> WindyWord -> WindyWordMobileApp.
The SHA-256 above must match the lab's. If it does not, this copy has drifted.
Clinic patient file: sneakyfree/Windy-Clinic -> stt-models/windy-lite.json
Licence
Upstream base is apache-2.0 on Hugging Face, verified against the source repo we converted from. Note OpenAI's Whisper GitHub repo states MIT while its HF model cards state Apache-2.0; we follow the HF distribution point, since that is what we actually pulled. Windy weights are a derivative and are distributed under the same terms, with attribution upstream.