AI Gateway · Component · realtime-audio-proxy
Realtime audio proxy and provider session integration
What it does
One gateway variant proxies bidirectional realtime sessions with lifecycle/usage accounting. Another builds speech-provider WebSocket descriptors and initial messages; it does not manage the complete voice-agent session.
Core pattern
Bidirectional audio event proxy + session state and usage
- Input
- Realtime session events/audio
- Output
- Realtime events and settlement records
Key parameters
| Name | Type | Default | Values |
|---|---|---|---|
| provider | string | — | — |
| voice_id | — | — | — |
| language | — | — | — |
| audio_encoding | string | pcm16 | pcm16 · g711_ulaw · g711_alaw |
| sample_rate_hz | integer | 24000 | 8000–192000 |
Maturity and limits
Realtime service, pricing and crash-test fixtures; Source paths were inspected statically; tests and live integrations were not executed. This rating describes implemented code, not production validation.
Credential-bearing configuration is excluded from this inventory. MIT; retain notices. Provider compatibility is configuration-dependent; metadata/privacy and optional payload plugins need review. Static inspection only. Variant-specific dependencies, limits and weaker maturity are recorded under variants; these implementations are alternatives, not one deployed system.
Used in pipelines
Industry tags
- HealthcareTransferable use
For organizing consented appointment-service interviews, use bidirectional audio event proxy followed by session state and usage.
Skills pack
Contract scaffold and deterministic mock
Download · 13KBSHA-256 40d9b1b145b774516064f51a3b7d6e958a5e86e4fe55b6480926cd59314736cf
Related components
-
Image generation, editing, upscaling and prediction jobs
Image generation or masked edits + async result tracking
Working 1 pipeline -
Speech transcription and audio translation adapters
Audio transcription or translation + language and format controls
Working 1 pipeline -
Asynchronous video generation and retained media
Async media requests + result polling and usage records
Working 1 pipeline -
Provider text summarization, sentiment and intent analysis
Supplied text topic and intent extraction + structured summaries
Working 2 pipelines