Interface: LiveSessionConfig

Defined in: packages/core/src/listen.ts:24

Configuration for starting a live transcription session

Properties

connectionTimeoutMs?

optional connectionTimeoutMs?: number

Defined in: packages/core/src/listen.ts:53

WebSocket connection timeout in ms (default: 10000)

---

encoding?

optional encoding?: "flac" | "pcm16" | "opus" | "mulaw"

Defined in: packages/core/src/listen.ts:28

Audio encoding (default: 'pcm16')

---

interimResults?

optional interimResults?: boolean

Defined in: packages/core/src/listen.ts:32

Return interim (partial) results (default: true)

---

language?

optional language?: string

Defined in: packages/core/src/listen.ts:30

BCP-47 language code (default: 'en')

---

model?

optional model?: string | ModelConfig

Defined in: packages/core/src/listen.ts:44

Override transcription provider (e.g., 'deepgram:nova-3') independent of bot's LLM model

---

onError?

optional onError?: (error) => void

Defined in: packages/core/src/listen.ts:51

Parameters

error

Error

Returns

void

---

onSpeechEnded?

optional onSpeechEnded?: () => void

Defined in: packages/core/src/listen.ts:50

Returns

void

---

onSpeechStarted?

optional onSpeechStarted?: () => void

Defined in: packages/core/src/listen.ts:49

Returns

void

---

onTranscript?

optional onTranscript?: (event) => void

Defined in: packages/core/src/listen.ts:48

Direct event callbacks (alternative to subscribeToAll)

Parameters

event

TranscriptEvent

Returns

void

---

providerConfig?

optional providerConfig?: ProviderConfig

Defined in: packages/core/src/listen.ts:46

Provider configuration (API keys, etc.)

---

punctuate?

optional punctuate?: boolean

Defined in: packages/core/src/listen.ts:34

Auto-punctuate (default: true)

---

sampleRate?

optional sampleRate?: number

Defined in: packages/core/src/listen.ts:26

Audio sample rate in Hz (default: 16000)

---

vad?

optional vad?: object

Defined in: packages/core/src/listen.ts:36

Voice Activity Detection config

enabled?

optional enabled?: boolean

silenceDurationMs?

optional silenceDurationMs?: number

Silence duration in ms before end-of-speech (default: 500)

threshold?

optional threshold?: number

VAD sensitivity threshold 0-1 (default: 0.5)