@plumbus/voice-livekit 0.1.2 → 0.1.3
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/CHANGELOG.md +11 -0
- package/README.md +1 -0
- package/dist/.tsbuildinfo +1 -1
- package/dist/client/client-noise-cancellation.d.ts +1 -0
- package/dist/client/client-noise-cancellation.d.ts.map +1 -1
- package/dist/client/client-noise-cancellation.js +4 -0
- package/dist/client/client-noise-cancellation.js.map +1 -1
- package/dist/client/livekit-session.d.ts.map +1 -1
- package/dist/client/livekit-session.js +7 -1
- package/dist/client/livekit-session.js.map +1 -1
- package/dist/runtime/livekit-agent-worker.d.ts +4 -0
- package/dist/runtime/livekit-agent-worker.d.ts.map +1 -1
- package/dist/runtime/livekit-agent-worker.js +10 -2
- package/dist/runtime/livekit-agent-worker.js.map +1 -1
- package/instructions/agent-worker.md +7 -0
- package/instructions/client-session.md +6 -0
- package/instructions/framework.md +1 -0
- package/package.json +3 -3
- package/dist/runtime/optional-cloud-registrations.d.ts +0 -12
- package/dist/runtime/optional-cloud-registrations.d.ts.map +0 -1
- package/dist/runtime/optional-cloud-registrations.js +0 -67
- package/dist/runtime/optional-cloud-registrations.js.map +0 -1
package/CHANGELOG.md
CHANGED
|
@@ -1,5 +1,16 @@
|
|
|
1
1
|
# Changelog
|
|
2
2
|
|
|
3
|
+
## 0.1.3
|
|
4
|
+
|
|
5
|
+
### Changed
|
|
6
|
+
|
|
7
|
+
- **Mic capture tuned for STT input.** `micConstraintsForNoiseCancellation` now explicitly sets `voiceIsolation: false` (livekit-client defaults it to true — a hidden extra enhancement stage in front of STT input); the mic publish disables Opus DTX (silence suppression can swallow quiet speakers' soft speech onsets) and enables RED for loss resilience.
|
|
8
|
+
|
|
9
|
+
### Fixed
|
|
10
|
+
|
|
11
|
+
- **`parsePcmFormat` understands short-form audio formats** (`pcm16-16k` / `pcm16-24k` / `pcm16-48k`) in the agent worker instead of silently defaulting to 16 kHz — which was correct only by coincidence for `pcm16-16k`.
|
|
12
|
+
|
|
13
|
+
|
|
3
14
|
## 0.1.2
|
|
4
15
|
|
|
5
16
|
### Fixed
|
package/README.md
CHANGED
|
@@ -128,6 +128,7 @@ import { createLiveKitVoiceSession } from '@plumbus/voice-livekit/client';
|
|
|
128
128
|
- `node_modules/@plumbus/voice-livekit/instructions/client-session.md` — browser session
|
|
129
129
|
- `node_modules/@plumbus/voice-livekit/instructions/agent-worker.md` — worker / CLI
|
|
130
130
|
- `node_modules/@plumbus/voice-livekit/instructions/noise-cancellation.md` — NC matrix
|
|
131
|
+
- `node_modules/@plumbus/voice/instructions/continuous-sessions.md` — talk-over re-queue, stitched transcripts, sentence chunker
|
|
131
132
|
|
|
132
133
|
## The Plumbus ecosystem
|
|
133
134
|
|