@jambonz/sdk
    Preparing search index...

    Interface Recognizer

    Speech-to-text recognition configuration.

    interface Recognizer {
        vendor: string;
        label?: string;
        language?: string;
        fallbackVendor?: string;
        fallbackLabel?: string;
        fallbackLanguage?: string;
        vad?: Vad;
        hints?: string[];
        hintsBoost?: number;
        altLanguages?: string[];
        profanityFilter?: boolean;
        interim?: boolean;
        singleUtterance?: boolean;
        dualChannel?: boolean;
        separateRecognitionPerChannel?: boolean;
        punctuation?: boolean;
        enhancedModel?: boolean;
        words?: boolean;
        diarization?: boolean;
        diarizationMinSpeakers?: number;
        diarizationMaxSpeakers?: number;
        interactionType?:
            | "unspecified"
            | "discussion"
            | "presentation"
            | "phone_call"
            | "voicemail"
            | "voice_search"
            | "voice_command"
            | "dictation";
        naicsCode?: number;
        identifyChannels?: boolean;
        vocabularyName?: string;
        vocabularyFilterName?: string;
        filterMethod?: "remove"
        | "mask"
        | "tag";
        model?: string;
        outputFormat?: "simple" | "detailed";
        profanityOption?: "masked" | "removed" | "raw";
        requestSnr?: boolean;
        initialSpeechTimeoutMs?: number;
        azureServiceEndpoint?: string;
        azureSttEndpointId?: string;
        asrDtmfTerminationDigit?: string;
        asrTimeout?: number;
        fastRecognitionTimeout?: number;
        minConfidence?: number;
        deepgramOptions?: Record<string, unknown>;
        googleOptions?: Record<string, unknown>;
        awsOptions?: Record<string, unknown>;
        azureOptions?: Record<string, unknown>;
        nuanceOptions?: Record<string, unknown>;
        ibmOptions?: Record<string, unknown>;
        nvidiaOptions?: Record<string, unknown>;
        sonioxOptions?: Record<string, unknown>;
        cobaltOptions?: Record<string, unknown>;
        assemblyAiOptions?: Record<string, unknown>;
        speechmaticsOptions?: Record<string, unknown>;
        openaiOptions?: Record<string, unknown>;
        houndifyOptions?: Record<string, unknown>;
        gladiaOptions?: Record<string, unknown>;
        elevenlabsOptions?: Record<string, unknown>;
        verbioOptions?: Record<string, unknown>;
        customOptions?: Record<string, unknown>;
    }
    Index

    Properties

    vendor label? language? fallbackVendor? fallbackLabel? fallbackLanguage? vad? hints? hintsBoost? altLanguages? profanityFilter? interim? singleUtterance? dualChannel? separateRecognitionPerChannel? punctuation? enhancedModel? words? diarization? diarizationMinSpeakers? diarizationMaxSpeakers? interactionType? naicsCode? identifyChannels? vocabularyName? vocabularyFilterName? filterMethod? model? outputFormat? profanityOption? requestSnr? initialSpeechTimeoutMs? azureServiceEndpoint? azureSttEndpointId? asrDtmfTerminationDigit? asrTimeout? fastRecognitionTimeout? minConfidence? deepgramOptions? googleOptions? awsOptions? azureOptions? nuanceOptions? ibmOptions? nvidiaOptions? sonioxOptions? cobaltOptions? assemblyAiOptions? speechmaticsOptions? openaiOptions? houndifyOptions? gladiaOptions? elevenlabsOptions? verbioOptions? customOptions?

    Properties

    vendor: string

    The STT vendor to use (e.g. 'google', 'aws', 'deepgram', 'microsoft').

    label?: string

    Label identifying a specific credential set for this vendor.

    language?: string

    Language code in BCP-47 format.

    fallbackVendor?: string

    Backup STT vendor if the primary fails.

    fallbackLabel?: string

    Credential label for the fallback vendor.

    fallbackLanguage?: string

    Language code for the fallback vendor.

    vad?: Vad

    Voice activity detection settings.

    hints?: string[]

    Words or phrases to boost recognition for.

    hintsBoost?: number

    Boost factor for hint words.

    altLanguages?: string[]

    Additional languages to listen for simultaneously.

    profanityFilter?: boolean

    Filter profanity from results.

    interim?: boolean

    Return interim (partial) results.

    singleUtterance?: boolean

    Stop after first complete utterance.

    dualChannel?: boolean

    Send separate audio channels for each call leg.

    separateRecognitionPerChannel?: boolean

    Independent recognition per audio channel.

    punctuation?: boolean

    Enable automatic punctuation.

    enhancedModel?: boolean

    Use enhanced (premium) recognition model.

    words?: boolean

    Include word-level timing.

    diarization?: boolean

    Enable speaker diarization.

    diarizationMinSpeakers?: number

    Minimum speakers expected for diarization.

    diarizationMaxSpeakers?: number

    Maximum speakers expected for diarization.

    interactionType?:
        | "unspecified"
        | "discussion"
        | "presentation"
        | "phone_call"
        | "voicemail"
        | "voice_search"
        | "voice_command"
        | "dictation"

    Hint about interaction type.

    naicsCode?: number

    NAICS industry code for domain-specific accuracy.

    identifyChannels?: boolean

    Identify and label which channel each segment came from.

    vocabularyName?: string

    Custom vocabulary resource name.

    vocabularyFilterName?: string

    Vocabulary filter name for masking/removing words.

    filterMethod?: "remove" | "mask" | "tag"

    How filtered words are handled.

    model?: string

    Specific recognition model name.

    outputFormat?: "simple" | "detailed"

    Level of detail in results.

    profanityOption?: "masked" | "removed" | "raw"

    How profanity is handled in results.

    requestSnr?: boolean

    Request signal-to-noise ratio info.

    initialSpeechTimeoutMs?: number

    Milliseconds to wait for initial speech.

    azureServiceEndpoint?: string

    Custom Azure Speech Services endpoint URL.

    azureSttEndpointId?: string

    Azure custom speech endpoint ID.

    asrDtmfTerminationDigit?: string

    DTMF digit that terminates recognition.

    asrTimeout?: number

    Max seconds for a complete recognition result.

    fastRecognitionTimeout?: number

    Timeout for fast recognition mode.

    minConfidence?: number

    Minimum confidence score (0-1) to accept a result.

    deepgramOptions?: Record<string, unknown>
    googleOptions?: Record<string, unknown>
    awsOptions?: Record<string, unknown>
    azureOptions?: Record<string, unknown>
    nuanceOptions?: Record<string, unknown>
    ibmOptions?: Record<string, unknown>
    nvidiaOptions?: Record<string, unknown>
    sonioxOptions?: Record<string, unknown>
    cobaltOptions?: Record<string, unknown>
    assemblyAiOptions?: Record<string, unknown>
    speechmaticsOptions?: Record<string, unknown>
    openaiOptions?: Record<string, unknown>
    houndifyOptions?: Record<string, unknown>
    gladiaOptions?: Record<string, unknown>
    elevenlabsOptions?: Record<string, unknown>
    verbioOptions?: Record<string, unknown>
    customOptions?: Record<string, unknown>

    Settings

    Member Visibility

    On This Page

    Properties