Interface STTOptions

interface STTOptions {
    agentContext?: string;
    agentContextCarryover?: boolean;
    apiKey?: string;
    baseUrl: string;
    bufferSizeMs: number;
    continuousPartials?: boolean;
    domain?: string;
    encoding: STTEncoding;
    endOfTurnConfidenceThreshold?: number;
    filterProfanity?: boolean;
    formatTurns?: boolean;
    inactivityTimeout?: number;
    includePartialTurns?: boolean;
    interruptionDelay?: number;
    keytermsPrompt?: string[];
    languageCodes?: string | string[];
    languageDetection?: boolean;
    maxSpeakers?: number;
    maxTurnSilence?: number;
    minTurnSilence?: number;
    mode?: "min_latency" | "balanced" | "max_accuracy";
    previousContextNTurns?: number;
    prompt?: string;
    redactPii?: boolean;
    redactPiiPolicies?: string[];
    redactPiiSub?: PiiRedactionSubstitution;
    sampleRate: number;
    speakerLabels?: boolean;
    speechModel: STTModels;
    vadThreshold?: number;
    voiceFocus?: VoiceFocus;
    voiceFocusThreshold?: number;
}

Properties

agentContext? agentContextCarryover? apiKey? baseUrl bufferSizeMs continuousPartials? domain? encoding endOfTurnConfidenceThreshold? filterProfanity? formatTurns? inactivityTimeout? includePartialTurns? interruptionDelay? keytermsPrompt? languageCodes? languageDetection? maxSpeakers? maxTurnSilence? minTurnSilence? mode? previousContextNTurns? prompt? redactPii? redactPiiPolicies? redactPiiSub? sampleRate speakerLabels? speechModel vadThreshold? voiceFocus? voiceFocusThreshold?

Properties

Optional agentContext

agentContext?: string

Only supported with the Universal-3 Pro model family.

Optional agentContextCarryover

agentContextCarryover?: boolean

When the model supports it, let an AgentSession push each assistant reply into agentContext so it is carried into the model's conversation context. Defaults to true for Universal-3 Pro models; set false to disable. Prior user turns are carried automatically by the model regardless of this flag. Ignored on models without context support.

Optional apiKey

apiKey?: string

baseUrl

baseUrl: string

bufferSizeMs

bufferSizeMs: number

How large each chunk of audio is before being sent to AssemblyAI, in milliseconds. Corresponds to Python's buffer_size_seconds (seconds there, ms here per this repo's time-unit convention).

Optional continuousPartials

continuousPartials?: boolean

Emit additional partial transcripts during long turns. Universal-3 Pro only.

Optional domain

domain?: string

encoding

encoding: STTEncoding

Optional endOfTurnConfidenceThreshold

endOfTurnConfidenceThreshold?: number

Optional filterProfanity

filterProfanity?: boolean

Filter profanity from transcripts. Connect-time only.

Optional formatTurns

formatTurns?: boolean

Optional inactivityTimeout

inactivityTimeout?: number

Session inactivity timeout in seconds. AssemblyAI accepts integer values from 5 to 3600; when unset, no inactivity timeout is applied.

Optional includePartialTurns

includePartialTurns?: boolean

Emit partial (non-final) turns. AssemblyAI defaults this to true, or false when redactPii is enabled. Connect-time only.

Optional interruptionDelay

interruptionDelay?: number

Delay before the first early partial is emitted, in milliseconds. Universal-3 Pro only.

Optional keytermsPrompt

keytermsPrompt?: string[]

Optional languageCodes

languageCodes?: string | string[]

Languages to steer transcription toward. Universal-3 Pro only.

Optional languageDetection

languageDetection?: boolean

Optional maxSpeakers

maxSpeakers?: number

Optional maxTurnSilence

maxTurnSilence?: number

Maximum silence (ms) before end-of-turn is forced regardless of confidence.

Optional minTurnSilence

minTurnSilence?: number

Minimum silence (ms) before a confident end-of-turn is finalized.

Optional mode

mode?: "min_latency" | "balanced" | "max_accuracy"

Accuracy/latency preset for the Universal-3 Pro model family: min_latency, balanced, or max_accuracy. Explicit turn-silence values still take precedence over mode defaults.

Optional previousContextNTurns

previousContextNTurns?: number

Only supported with the Universal-3 Pro model family. Set at connection time only.

Optional prompt

prompt?: string

Only supported with the Universal-3 Pro model family.

Optional redactPii

redactPii?: boolean

Redact PII from final turns. AssemblyAI disables partial turns when this is enabled unless includePartialTurns is set; partials are never redacted. Connect-time only.

Optional redactPiiPolicies

redactPiiPolicies?: string[]

PII policies to redact, such as person_name or phone_number. Redacts all detected PII when unset. Requires redactPii. Connect-time only.

Optional redactPiiSub

redactPiiSub?: PiiRedactionSubstitution

Replacement for redacted PII: hash (AssemblyAI default) or entity_name, which writes [PERSON_NAME]-style placeholders. Requires redactPii. Connect-time only.

sampleRate

sampleRate: number

Optional speakerLabels

speakerLabels?: boolean

Enable speaker diarization; the turn's speaker label is surfaced as speakerId.

speechModel

speechModel: STTModels

Model selected at construction. Create a new STT to switch models.

Optional vadThreshold

vadThreshold?: number

Optional voiceFocus

voiceFocus?: VoiceFocus

Isolate the primary voice and suppress background noise. Connect-time only.

Optional voiceFocusThreshold

voiceFocusThreshold?: number

Background audio suppression aggressiveness, from 0.0 to 1.0. Connect-time only.

Settings

Member Visibility

Theme

On This Page

agentContextagentContextCarryoverapiKeybaseUrlbufferSizeMscontinuousPartialsdomainencodingendOfTurnConfidenceThresholdfilterProfanityformatTurnsinactivityTimeoutincludePartialTurnsinterruptionDelaykeytermsPromptlanguageCodeslanguageDetectionmaxSpeakersmaxTurnSilenceminTurnSilencemodepreviousContextNTurnspromptredactPiiredactPiiPoliciesredactPiiSubsampleRatespeakerLabelsspeechModelvadThresholdvoiceFocusvoiceFocusThreshold