ryu-stt 0.1.2

Speech-to-text (STT) modality primitive for Ryu: `transcribe(audio) -> text` with a swappable engine seam (in-process parakeet ONNX by default; whisper.cpp and Gateway-routed cloud Whisper as HTTP proxies). An extracted Core capability crate — in-process by default and consumed as a NON-optional path dependency (the voice/meetings/hardware data paths reach it unconditionally). The parakeet ONNX inference is the genuinely in-process hot path (feature `voice-parakeet`); whisper + gateway engines are thin HTTP proxies. Host couplings the crate cannot own (whisper base-url, Gateway url/bearer, the parakeet model dir) are injected via the narrow `SttHost` trait, so the crate has ZERO dependency on apps/core.
Documentation

Builds

ryu-stt's sandbox limits

All the builds on docs.rs are executed inside a sandbox with limited resources. The limits for this crate are the following:

Available RAM 6.44 GB
Maximum rustdoc execution time 15m
Maximum size of a build log 102.4 kB
Network access blocked
Maximum number of build targets 10

If a build fails because it hit one of those limits please open an issue to get them increased.