Repository navigation
Captions pack 2 (speech models for more stream languages)
Pre-releaseOptional speech models for captions in more stream languages in PurpleTV ReVive (ad break captions and live captions). You do not need to download anything here by hand: pick the streamer's language under PurpleTV › Player › During ad breaks › Stream language (or under PurpleTV › Translation), then tap "Download speech model". The module checks every file against its pinned size and SHA-256 and only then uses it. Everything runs on your phone, and no audio or text is sent anywhere.
The captions engine (sherpa-onnx) and the English model stay in captions-pack-1; you need that engine too.
This is a pre-release on purpose, so it never shows up as the module's latest version.
What is in here
| File | Stream language | Download | On the phone |
|---|---|---|---|
purpletv-captions-model-kroko-es-2025-08-06.zip |
Spanish | 120 MB | 156 MB |
purpletv-captions-model-kroko-de-2025-08-06.zip |
German | 55 MB | 71 MB |
purpletv-captions-model-kroko-fr-2025-08-06.zip |
French | 55 MB | 71 MB |
purpletv-captions-model-vosk-small-ru-2025-08-16.zip |
Russian | 23 MB | 29 MB |
purpletv-captions-model-zipformer-ko-2024-06-16.zip |
Korean | 123 MB | 141 MB |
purpletv-captions-model-zipformer-ctc-zh-small-2025-04-01.zip |
Chinese (Mandarin) | 20 MB | 26 MB |
Each zip holds the model files exactly as published in the sherpa-onnx copies linked below (nothing changed or retrained), plus a LICENSE.txt and a README.txt with the credits, the source commit and every file's SHA-256. The Apache-2.0 models also carry the full licence text.
Credits and licences
- Kroko ASR by Banafo (Spanish, German and French), Creative Commons Attribution-ShareAlike (CC-BY-SA, current version 4.0: https://creativecommons.org/licenses/by-sa/4.0/): https://huggingface.co/Banafo/Kroko-ASR. Converted for sherpa-onnx by csukuangfj, for example https://huggingface.co/csukuangfj/sherpa-onnx-streaming-zipformer-es-kroko-2025-08-06. If you share these files, keep the credit and the same licence.
- Vosk small streaming Russian model by Alpha Cephei, Apache-2.0: https://huggingface.co/alphacep/vosk-model-small-streaming-ru. Converted for sherpa-onnx by csukuangfj: https://huggingface.co/csukuangfj/sherpa-onnx-streaming-zipformer-small-ru-vosk-int8-2025-08-16
- Korean streaming Zipformer by johnBamma, trained with icefall, Apache-2.0: https://huggingface.co/johnBamma/icefall-asr-ksponspeech-pruned-transducer-stateless7-streaming-2024-06-12. Converted for sherpa-onnx by k2-fsa: https://huggingface.co/k2-fsa/sherpa-onnx-streaming-zipformer-korean-2024-06-16
- Small streaming Zipformer CTC Chinese model by k2-fsa (csukuangfj), Apache-2.0: https://huggingface.co/csukuangfj/icefall-streaming-zipformer-small-ctc-zh-2025-04-01 and https://huggingface.co/csukuangfj/sherpa-onnx-streaming-zipformer-small-ctc-zh-int8-2025-04-01
- sherpa-onnx by k2-fsa, Apache-2.0, runs them all: https://github.com/k2-fsa/sherpa-onnx
Thank you to everyone behind these projects!