Voice WebSocket transport
Voice
Voice WebSocket transport
Upgrades to WebSocket carrying frames per voice-session-transport-spec.
Auth via vsess_* HMAC token (60s TTL) on ?auth= query. Optional
?sample_rate=N (8000/16000/24000/32000/44100/48000) enables
bidirectional resampling for clients that can’t natively produce
16 kHz mic audio or consume 24 kHz playback.
GET
Voice WebSocket transport
Path Parameters
Query Parameters
Single-use HMAC token from POST /v1/runtime/voice/sessions/bootstrap (user-api).
Optional sample-rate override; gateway resamples upstream + downstream PCM.
Available options:
8000, 16000, 24000, 32000, 44100, 48000 Response
WebSocket upgrade.