@garrix82/reactgenie-lib 1.3.0

package/dist/speech-recognition/speech-recognition-unified.d.ts

@@ -0,0 +1,93 @@
+import React from "react";
+import { isGroqSpeechAvailable, getGroqSpeechCapabilities } from "./speech-recognition-groq";
+import { type LocalWhisperMode } from "./speech-recognition-mlx";
+import { isOpenAISpeechAvailable, getOpenAISpeechCapabilities } from "./speech-recognition-openai";
+export interface UnifiedSpeechRecognizerProps {
+    shouldListen: boolean;
+    speechStatusCallback: (status: boolean, transcript: string) => void;
+    speechResultCallback: (result: string) => void;
+    speechTranslationCallback?: (translation: string) => void;
+    speechReadyCallback?: (ready: boolean) => void;
+    waveformPreviewCallback?: (waveform: number[]) => void;
+    clientSecret?: string;
+    language?: string;
+    onError?: (error: Error) => void;
+    requestContextProvider?: () => {
+        threadId?: string | null;
+        turnId?: string | null;
+        sessionId?: string | number | null;
+        sentryTrace?: string | null;
+        baggage?: string | null;
+    } | null | undefined;
+    config?: SpeechRecognitionConfig;
+}
+export interface GroqConfigProps {
+    enabled?: boolean;
+    apiKey?: string;
+    apiBaseUrl?: string;
+    transcriptionEndpoint?: string;
+    stream?: boolean;
+}
+export interface OpenAIConfigProps {
+    proxyBaseUrl?: string;
+    transcriptionSessionEndpoint?: string;
+    translationEndpoint?: string;
+    realtimeWebRtcUrl?: string;
+    realtimeModel?: "gpt-4o-transcribe" | "gpt-4o-mini-transcribe";
+    translationModel?: string;
+    vadEnabled?: boolean;
+    autoStopOnVadSilence?: boolean;
+    vadThreshold?: number;
+    vadPrefixPaddingMs?: number;
+    vadSilenceDurationMs?: number;
+    vadIdleTimeoutMs?: number;
+    prewarmOnMount?: boolean;
+}
+export interface LocalModelConfigProps {
+    enabledOnWeb?: boolean;
+    baseUrl?: string;
+    mode?: LocalWhisperMode;
+    model?: string;
+}
+export interface SpeechRecognitionConfig {
+    provider?: "groq" | "openai";
+    fallbackToWebSpeech?: boolean;
+    groq?: GroqConfigProps;
+    openai?: OpenAIConfigProps;
+    localModel?: LocalModelConfigProps;
+}
+export declare function getSpeechRecognitionConfig(overrides?: Partial<SpeechRecognitionConfig>): SpeechRecognitionConfig;
+declare function checkWebSpeechAvailability(): boolean;
+export declare const isWebSpeechAvailable: typeof checkWebSpeechAvailability;
+type SpeechRecognitionMode = "openai-web" | "openai-native" | "groq-web" | "groq-native" | "webspeech" | "mlx-local" | "none";
+export declare function selectSpeechRecognitionMethod(config: SpeechRecognitionConfig, clientSecret?: string): SpeechRecognitionMode;
+export declare const UnifiedSpeechRecognizer: React.FC<UnifiedSpeechRecognizerProps>;
+export declare function getSpeechRecognitionStatus(overrides?: Partial<SpeechRecognitionConfig>): {
+    method: SpeechRecognitionMode;
+    config: SpeechRecognitionConfig;
+    capabilities: {
+        groq: {
+            available: boolean;
+            supportedMimeTypes: string[];
+            platform: "ios" | "android" | "windows" | "macos" | "web";
+        };
+        openai: {
+            available: boolean;
+            platform: "ios" | "android" | "windows" | "macos" | "web";
+            supportsServerVAD: boolean;
+            transport: string;
+        };
+        webSpeech: {
+            available: boolean;
+            platform: "ios" | "android" | "windows" | "macos" | "web";
+        };
+        localModel: {
+            enabled: boolean;
+            url: string;
+            mode: LocalWhisperMode;
+        };
+    };
+    recommendations: string[];
+};
+export { isGroqSpeechAvailable, getGroqSpeechCapabilities, isOpenAISpeechAvailable, getOpenAISpeechCapabilities, };
+export default UnifiedSpeechRecognizer;