stt-vocab.yaml 1.4 KB

1234567891011121314151617181920212223242526272829
  1. name: stt-vocab
  2. repo: https://github.com/pierrenode/hermes-stt-vocab
  3. sha: 54e83f9289315af6d47c7c6fdeb5714a08b35262
  4. subdir: stt-vocab
  5. description: >-
  6. Puts the names a speech-to-text model mis-hears into the transcription prompt Hermes sends with
  7. each voice message: terms you list (globally or per surface) and names found in this profile's
  8. memories/USER.md and MEMORY.md, most important last, within the 896 characters Hermes keeps for
  9. Whisper-family backends; your stt.prompt stays in front. Notes written in German, Spanish,
  10. French, Italian, Dutch, Portuguese or Turkish can be named in the languages setting.
  11. Disclosure — register() adds one pre_transcription hook that returns a new prompt string;
  12. memory-derived names go only to local backends (faster-whisper, local_command) unless
  13. allow_remote is set, and the memory files are not read otherwise; reads at most 256 KB per
  14. memory file; no tools, network access, subprocesses or writes; Hermes uploads the prompt with
  15. the audio to the configured STT provider.
  16. maintainer: pierrenode
  17. tier: community
  18. category: voice
  19. requires_hermes: ">=0.21"
  20. docs_url: https://github.com/pierrenode/hermes-stt-vocab
  21. version: "1.1.0"
  22. image: https://raw.githubusercontent.com/pierrenode/hermes-stt-vocab/6145d065d50f196d51b3ab80fe4a76fb2827c115/docs/card.png
  23. platforms: []
  24. capabilities:
  25. provides_tools: []
  26. provides_hooks:
  27. - pre_transcription
  28. provides_middleware: []
  29. requires_env: []