Circuit-Forge/circuitforge-core
3
0
Fork 0
Code Issues 2 Pull requests Projects Releases 6 Packages Wiki Activity Actions 6
circuitforge-core/docs/modules/stt.md
pyr0ball 383897f990
Some checks are pending
CI / test (push) Waiting to run
Details
Mirror / mirror (push) Waiting to run
Details
Release — PyPI / release (push) Waiting to run
Details
feat: platforms module + docs + scripts 
- platforms/: eBay platform adapter (snipe integration layer)
- docs/: developer guide, module reference, getting-started docs
- scripts/: utility scripts for development and deployment
2026-04-24 15:23:16 -07:00

1 KiB
Raw Blame History

stt

Speech-to-text router. Stub — not yet implemented.

from circuitforge_core.stt import STTRouter  # planned

Planned design

The STT module will provide a unified interface over local speech-to-text backends, following the same fallback-chain pattern as the LLM router.

Planned backends:

  • whisper_cpp — local, CPU/GPU, various model sizes
  • faster_whisper — local, GPU-accelerated, CTranslate2 backend
  • whisper_openai — cloud, requires OPENAI_API_KEY

Planned use cases across the menagerie:

  • Osprey: transcribe hold music + IVR menu audio for navigation
  • Linnet: real-time speech annotation (tone classification requires transcript)
  • Peregrine: interview practice sessions

Current status

The circuitforge_core.stt directory exists in-tree with a stub __init__.py. No working implementation yet. Planned for the milestone after Osprey reaches beta.

If you need STT before this module ships, use faster-whisper directly in the product and plan to migrate to this interface once it stabilizes.