@picovoice/leopard-web/dist/types/leopard.d.ts

Leopard Speech-to-Text engine for web browsers (via WebAssembly)

import { LeopardModel, LeopardOptions, LeopardTranscript } from './types';
export declare class Leopard {
    private _module?;
    private readonly _pv_leopard_process;
    private readonly _pv_leopard_delete;
    private readonly _sampleRate;
    private readonly _version;
    private readonly _processMutex;
    private readonly _objectAddress;
    private readonly _transcriptAddressAddress;
    private readonly _numWordsAddress;
    private readonly _wordsAddressAddress;
    private readonly _messageStackAddressAddressAddress;
    private readonly _messageStackDepthAddress;
    private static _wasmSimd;
    private static _wasmSimdLib;
    private static _wasmPThread;
    private static _wasmPThreadLib;
    private static _sdk;
    private static _leopardMutex;
    private constructor();
    /**
     * Get Leopard engine version.
     */
    get version(): string;
    /**
     * Get sample rate.
     */
    get sampleRate(): number;
    /**
     * Set base64 wasm file with SIMD feature.
     * @param wasmSimd Base64'd wasm file to use to initialize wasm.
     */
    static setWasmSimd(wasmSimd: string): void;
    /**
     * Set base64 SIMD wasm file in text format.
     * @param wasmSimdLib Base64'd wasm file in text format.
     */
    static setWasmSimdLib(wasmSimdLib: string): void;
    /**
     * Set base64 wasm file with SIMD and pthread feature.
     * @param wasmPThread Base64'd wasm file to use to initialize wasm.
     */
    static setWasmPThread(wasmPThread: string): void;
    /**
     * Set base64 SIMD and thread wasm file in text format.
     * @param wasmPThreadLib Base64'd wasm file in text format.
     */
    static setWasmPThreadLib(wasmPThreadLib: string): void;
    static setSdk(sdk: string): void;
    /**
     * Creates an instance of the Picovoice Leopard Speech-to-Text engine.
     * Behind the scenes, it requires the WebAssembly code to load and initialize before
     * it can create an instance.
     *
     * @param accessKey AccessKey obtained from Picovoice Console (https://console.picovoice.ai/)
     * @param model Leopard model options.
     * @param model.base64 The model in base64 string to initialize Leopard.
     * @param model.publicPath The model path relative to the public directory.
     * @param model.customWritePath Custom path to save the model in storage.
     * Set to a different name to use multiple models across `leopard` instances.
     * @param model.forceWrite Flag to overwrite the model in storage even if it exists.
     * @param model.version Version of the model file. Increment to update the model file in storage.
     * @param options Optional arguments.
     * @param options.device String representation of the device (e.g., CPU or GPU) to use. If set to `best`, the most
     * suitable device is selected automatically. If set to `gpu`, the engine uses the first available GPU device. To select a specific
     * GPU device, set this argument to `gpu:${GPU_INDEX}`, where `${GPU_INDEX}` is the index of the target GPU. If set to
     * `cpu`, the engine will run on the CPU with the default number of threads. To specify the number of threads, set this
     * argument to `cpu:${NUM_THREADS}`, where `${NUM_THREADS}` is the desired number of threads.
     * @param options.enableAutomaticPunctuation Flag to enable automatic punctuation insertion.
     * @param options.enableDiarization Flag to enable speaker diarization, which allows Leopard to differentiate speakers
     * as part of the transcription process. Word metadata will include a `speakerTag` to identify unique speakers.
     *
     * @returns An instance of the Leopard engine.
     */
    static create(accessKey: string, model: LeopardModel, options?: LeopardOptions): Promise<Leopard>;
    static _init(accessKey: string, modelPath: string, options?: LeopardOptions): Promise<Leopard>;
    /**
     * Processes audio. The required sample rate can be retrieved from '.sampleRate'. The audio needs to be
     * 16-bit linearly-encoded. Furthermore, the engine operates on single-channel audio.
     *
     * @param pcm A frame of audio with properties described above.
     * @return The transcript.
     */
    process(pcm: Int16Array): Promise<LeopardTranscript>;
    /**
     * Releases resources acquired by WebAssembly module.
     */
    release(): Promise<void>;
    private static initWasm;
    private static getMessageStack;
    private static wrapAsyncFunction;
}
//# sourceMappingURL=leopard.d.ts.map