Transaction-Science/open-standards
Watch
1
0
Fork
You've already forked open-standards
 0
Code Issues Pull requests Projects Releases 17 Packages Wiki Activity Actions 200
open-standards/aac/aac-rs/data/ai/LICENSES.md
Transaction Science f71fa28447 feat(aac sentence helper): 31 languages, each tested before it is offered 
41 more languages were tested on the 20 everyday exchanges, with Gemma 4
E2B and E4B. Two independent model readers read every answer, and Gemma 4
31B a third time wherever their verdicts could change the outcome (it
agreed with the hand reading of English, Spanish, German and Hindi on 75
of 80 verdicts). A language is ready on a model at 15 of 20 usable
sentences and 60 of 80 good replies, and is offered with the smallest
model it is ready on.

- Ready with E2B: Arabic, Danish, French, Indonesian, Korean, Malay,
  Persian, Portuguese, Russian, Swedish, Turkish, Urdu, Vietnamese (with
  English, Spanish, German and Hindi: 17).
- Ready with E4B: Chinese, Dutch, Gujarati, Hebrew, Italian, Japanese,
  Marathi, Norwegian, Polish, Punjabi, Tagalog, Telugu, Thai, Ukrainian.
- Not yet, each with what went wrong: Amharic, Bengali, Czech, Finnish,
  Greek, Hausa, Hungarian, Kannada, Malayalam, Romanian, Swahili, Tamil,
  Yoruba, Zulu.

Every answer and verdict is in eval/results-2026-10-01-languages.json. No
native speaker has read these yet. The committed catalogue's test now
checks every ready language and the larger model's cover.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-10-01 18:39:17 -04:00

2 KiB
Raw Permalink Blame History

Sentence helper: licences

The sentence helper in AuraSense AAC (aac-web) downloads one language model, on request, and runs it on the person's device. Nothing they type is sent anywhere.

Gemma 4 E2B

  • Gemma 4 E2B by Google DeepMind, in Google AI Edge's web build for LiteRT-LM (litert-community on Hugging Face).
  • Licence: Apache-2.0 (https://ai.google.dev/gemma/docs/gemma_4_license); the text is licences/Apache-2.0.c71d239df917.txt.
  • Source: huggingface.co/litert-community/gemma-4-E2B-it-litert-lm@b3ca0d2f07/gemma-4-E2B-it-web.litertlm (2008432640 bytes, SHA-256 3a08e8d94e23b814ae5414469c370c503813949acb8ceaa17e4ebf8a35af35b5), unchanged, published here in 30 parts of at most 64 MiB, each named by its own SHA-256. Together they are the file.
  • Size: 2.0 GB. Ready in: Arabic (ar), Danish (da), German (de), English (en), Spanish (es), Persian (fa), French (fr), Hindi (hi), Indonesian (id), Korean (ko), Malay (ms), Portuguese (pt), Russian (ru), Swedish (sv), Turkish (tr), Urdu (ur), Vietnamese (vi).

Gemma 4 E4B

  • Gemma 4 E4B by Google DeepMind, in Google AI Edge's web build for LiteRT-LM (litert-community on Hugging Face).
  • Licence: Apache-2.0 (https://ai.google.dev/gemma/docs/gemma_4_license); the text is licences/Apache-2.0.c71d239df917.txt.
  • Source: huggingface.co/litert-community/gemma-4-E4B-it-litert-lm@2eee7ac325/gemma-4-E4B-it-web.litertlm (2969059328 bytes, SHA-256 3904d826d5dddd25ea173e85204caec09e68ba038116e9b992b69cbdc94f57a0), unchanged, published here in 45 parts of at most 64 MiB, each named by its own SHA-256. Together they are the file.
  • Size: 3.0 GB. Ready in: Gujarati (gu), Hebrew (he), Italian (it), Japanese (ja), Marathi (mr), Norwegian (nb), Dutch (nl), Punjabi (pa), Polish (pl), Telugu (te), Thai (th), Filipino (tl), Ukrainian (uk), Chinese (zh).

The runtime that runs it (LiteRT-LM, Apache-2.0) is vendored with its own licence and notice under vendor/litert-lm-<version>-<revision>/.