feat(speech): model-agnostic, local-first speech synthesis packages #23

Merged
crapougnax merged 2 commits from feat/69-speech-packages into develop 2026-10-27 15:46:10 +00:00
Owner

New packages @quatrain/speech, @quatrain/speech-browser, @quatrain/speech-api (0.1.0). 29 jest tests, tsc clean per package.

Closes #22

Migrated from GitHub https://github.com/Quatrain/Core/pull/70.

New packages `@quatrain/speech`, `@quatrain/speech-browser`, `@quatrain/speech-api` (0.1.0). 29 jest tests, tsc clean per package. Closes #22 > Migrated from GitHub https://github.com/Quatrain/Core/pull/70.
- @quatrain/speech: synthesizer contract (locale, voice, style, rate, pitch),
  sentence splitting, SpeechController (preference list, sentence-by-sentence
  playback with look-ahead, fallback before first audio, stop, normalized
  started/ended/interrupted events).
- @quatrain/speech-browser: on-device Web Speech adapter with voice selection.
- @quatrain/speech-http: plays audio from any HTTP endpoint, download starts at
  prepare time.

Closes #69
refactor(speech): rename speech-http to speech-api
Some checks failed
Quatrain Core Sovereign CI / Monorepo Build, Tests & Quality Audit (pull_request) Failing after 24s
Quatrain Core Sovereign CI / Monorepo Build, Tests & Quality Audit (push) Failing after 2m47s
8ae06091df
ApiSpeechSynthesizer (id 'api') plays audio returned by a speech API endpoint.

Refs #69
crapougnax merged commit fb21c7100c into develop 2026-10-27 15:46:10 +00:00
crapougnax deleted branch feat/69-speech-packages 2026-10-27 15:46:10 +00:00
Sign in to join this conversation.
No description provided.