@lokutor/sdk 1.1.9 → 1.1.11

package/dist/index.d.mts

@@ -119,14 +119,19 @@ declare class VoiceAgentClient {
     private messages;
     private visemeListeners;
     private wantVisemes;
-    private serverUrl;
+    private audioManager;
+    private enableAudio;
+    private isUserDisconnect;
+    private reconnecting;
+    private reconnectAttempts;
+    private maxReconnectAttempts;
     constructor(config: LokutorConfig & {
         prompt: string;
         voice?: VoiceStyle;
         language?: Language;
         visemes?: boolean;
-        serverUrl?: string;
         onVisemes?: (visemes: Viseme[]) => void;
+        enableAudio?: boolean;
     });
     /**
      * Connect to the Lokutor Voice Agent server
@@ -157,6 +162,15 @@ declare class VoiceAgentClient {
      * Disconnect from the server
      */
     disconnect(): void;
+    /**
+     * Toggles the microphone mute state (if managed by client)
+     * returns the new mute state
+     */
+    toggleMute(): boolean;
+    /**
+     * Gets the microphone volume amplitude 0-1 (if managed by client)
+     */
+    getAmplitude(): number;
     /**
      * Update the system prompt mid-conversation
      */
@@ -326,9 +340,10 @@ declare class BrowserAudioManager {
     private scriptProcessor;
     private analyserNode;
     private mediaStream;
+    private resampler;
     private nextPlaybackTime;
     private activeSources;
-    private audioClockOffset;
+    private playbackQueue;
     private inputSampleRate;
     private outputSampleRate;
     private autoGainControl;
@@ -338,7 +353,6 @@ declare class BrowserAudioManager {
     private onInputError?;
     private isMuted;
     private isListening;
-    private resampler;
     constructor(config?: BrowserAudioConfig);
     /**
      * Initialize the AudioContext and analyser
@@ -348,77 +362,60 @@ declare class BrowserAudioManager {
      * Start capturing audio from the microphone
      */
     startMicrophone(onAudioInput: (pcm16Data: Uint8Array) => void): Promise<void>;
+    /**
+     * Internal method to process microphone audio data
+     */
     private _processAudioInput;
+    /**
+     * Stop capturing microphone input
+     */
     stopMicrophone(): void;
     /**
      * Play back audio received from the server
+     * @param pcm16Data Int16 PCM audio data at SPEAKER_SAMPLE_RATE
      */
     playAudio(pcm16Data: Uint8Array): void;
-    private _schedulePlayback;
     /**
-     * Get the current high-precision audio clock offset for viseme synchronization.
-     * Total stream time (in ms) = (audioContext.currentTime - audioClockOffset) * 1000
+     * Internal method to schedule and play audio with sample-accurate timing
      */
-    getAudioClockOffset(): number | null;
+    private _schedulePlayback;
     /**
-     * Reset the audio clock offset (call when a response is interrupted or finished)
+     * Stop all currently playing audio and clear the queue
      */
-    resetAudioClock(): void;
     stopPlayback(): void;
+    /**
+     * Toggle mute state
+     */
     setMuted(muted: boolean): void;
-    isMicMuted(): boolean;
-    getAmplitude(): number;
-    getFrequencyData(): Uint8Array;
-    getWaveformData(): Uint8Array;
-    cleanup(): void;
-    getAudioContext(): AudioContext | null;
-}
-
-/**
- * High-level AI Voice Agent for browser-based conversations.
- *
- * This class orchestrates microphone input, AI processing, and
- * speaker output, providing a simple interface for building
- * voice assistants with lip-sync support.
- */
-declare class VoiceAgent {
-    private client;
-    private audioManager;
-    private options;
-    private isConnected;
-    private visemeQueue;
-    constructor(options: VoiceAgentOptions & {
-        apiKey: string;
-    });
     /**
-     * Initialize hardware and connect to the AI server.
-     * This must be called in response to a user guesture (like a click)
-     * to satisfy browser AudioContext requirements.
+     * Get current mute state
      */
-    connect(): Promise<boolean>;
+    isMicMuted(): boolean;
     /**
-     * Get the current amplitude/volume of the microphone or output audio.
-     * Useful for voice activity visualization.
-     * @returns value between 0 and 1
+     * Get current amplitude from analyser (for visualization)
+     * Returns value between 0 and 1
      */
     getAmplitude(): number;
     /**
-     * Mute or unmute the microphone.
+     * Get frequency data from analyser for visualization
      */
-    toggleMute(): boolean;
+    getFrequencyData(): Uint8Array;
+    /**
+     * Get time-domain data from analyser for waveform visualization
+     */
+    getWaveformData(): Uint8Array;
     /**
-     * High-precision method to get visemes that should be active
-     * at the current playback frame. Use this in a requestAnimationFrame loop.
+     * Cleanup and close AudioContext
      */
-    getFrameVisemes(): Viseme[];
+    cleanup(): void;
     /**
-     * Change the system prompt mid-conversation.
+     * Get current audio context state
      */
-    updatePrompt(newPrompt: string): void;
+    getState(): 'running' | 'suspended' | 'closed' | 'interrupted' | null;
     /**
-     * Disconnect and release audio resources.
+     * Check if microphone is currently listening
      */
-    disconnect(): void;
+    isRecording(): boolean;
 }
 
-export { AUDIO_CONFIG, type AnalyserConfig, type BrowserAudioConfig, BrowserAudioManager, type BrowserAudioOptions, DEFAULT_URLS, Language, type LokutorConfig, StreamResampler, type SynthesizeOptions, TTSClient, type Viseme, VoiceAgent, VoiceAgentClient, type VoiceAgentOptions, VoiceStyle, applyLowPassFilter, bytesToPcm16, calculateRMS, float32ToPcm16, normalizeAudio, pcm16ToBytes, pcm16ToFloat32, resample, resampleWithAntiAliasing, simpleConversation, simpleTTS };
+export { AUDIO_CONFIG, type AnalyserConfig, type BrowserAudioConfig, BrowserAudioManager, type BrowserAudioOptions, DEFAULT_URLS, Language, type LokutorConfig, StreamResampler, type SynthesizeOptions, TTSClient, type Viseme, VoiceAgentClient, type VoiceAgentOptions, VoiceStyle, applyLowPassFilter, bytesToPcm16, calculateRMS, float32ToPcm16, normalizeAudio, pcm16ToBytes, pcm16ToFloat32, resample, resampleWithAntiAliasing, simpleConversation, simpleTTS };

package/dist/index.d.ts

@@ -119,14 +119,19 @@ declare class VoiceAgentClient {
     private messages;
     private visemeListeners;
     private wantVisemes;
-    private serverUrl;
+    private audioManager;
+    private enableAudio;
+    private isUserDisconnect;
+    private reconnecting;
+    private reconnectAttempts;
+    private maxReconnectAttempts;
     constructor(config: LokutorConfig & {
         prompt: string;
         voice?: VoiceStyle;
         language?: Language;
         visemes?: boolean;
-        serverUrl?: string;
         onVisemes?: (visemes: Viseme[]) => void;
+        enableAudio?: boolean;
     });
     /**
      * Connect to the Lokutor Voice Agent server
@@ -157,6 +162,15 @@ declare class VoiceAgentClient {
      * Disconnect from the server
      */
     disconnect(): void;
+    /**
+     * Toggles the microphone mute state (if managed by client)
+     * returns the new mute state
+     */
+    toggleMute(): boolean;
+    /**
+     * Gets the microphone volume amplitude 0-1 (if managed by client)
+     */
+    getAmplitude(): number;
     /**
      * Update the system prompt mid-conversation
      */
@@ -326,9 +340,10 @@ declare class BrowserAudioManager {
     private scriptProcessor;
     private analyserNode;
     private mediaStream;
+    private resampler;
     private nextPlaybackTime;
     private activeSources;
-    private audioClockOffset;
+    private playbackQueue;
     private inputSampleRate;
     private outputSampleRate;
     private autoGainControl;
@@ -338,7 +353,6 @@ declare class BrowserAudioManager {
     private onInputError?;
     private isMuted;
     private isListening;
-    private resampler;
     constructor(config?: BrowserAudioConfig);
     /**
      * Initialize the AudioContext and analyser
@@ -348,77 +362,60 @@ declare class BrowserAudioManager {
      * Start capturing audio from the microphone
      */
     startMicrophone(onAudioInput: (pcm16Data: Uint8Array) => void): Promise<void>;
+    /**
+     * Internal method to process microphone audio data
+     */
     private _processAudioInput;
+    /**
+     * Stop capturing microphone input
+     */
     stopMicrophone(): void;
     /**
      * Play back audio received from the server
+     * @param pcm16Data Int16 PCM audio data at SPEAKER_SAMPLE_RATE
      */
     playAudio(pcm16Data: Uint8Array): void;
-    private _schedulePlayback;
     /**
-     * Get the current high-precision audio clock offset for viseme synchronization.
-     * Total stream time (in ms) = (audioContext.currentTime - audioClockOffset) * 1000
+     * Internal method to schedule and play audio with sample-accurate timing
      */
-    getAudioClockOffset(): number | null;
+    private _schedulePlayback;
     /**
-     * Reset the audio clock offset (call when a response is interrupted or finished)
+     * Stop all currently playing audio and clear the queue
      */
-    resetAudioClock(): void;
     stopPlayback(): void;
+    /**
+     * Toggle mute state
+     */
     setMuted(muted: boolean): void;
-    isMicMuted(): boolean;
-    getAmplitude(): number;
-    getFrequencyData(): Uint8Array;
-    getWaveformData(): Uint8Array;
-    cleanup(): void;
-    getAudioContext(): AudioContext | null;
-}
-
-/**
- * High-level AI Voice Agent for browser-based conversations.
- *
- * This class orchestrates microphone input, AI processing, and
- * speaker output, providing a simple interface for building
- * voice assistants with lip-sync support.
- */
-declare class VoiceAgent {
-    private client;
-    private audioManager;
-    private options;
-    private isConnected;
-    private visemeQueue;
-    constructor(options: VoiceAgentOptions & {
-        apiKey: string;
-    });
     /**
-     * Initialize hardware and connect to the AI server.
-     * This must be called in response to a user guesture (like a click)
-     * to satisfy browser AudioContext requirements.
+     * Get current mute state
      */
-    connect(): Promise<boolean>;
+    isMicMuted(): boolean;
     /**
-     * Get the current amplitude/volume of the microphone or output audio.
-     * Useful for voice activity visualization.
-     * @returns value between 0 and 1
+     * Get current amplitude from analyser (for visualization)
+     * Returns value between 0 and 1
      */
     getAmplitude(): number;
     /**
-     * Mute or unmute the microphone.
+     * Get frequency data from analyser for visualization
      */
-    toggleMute(): boolean;
+    getFrequencyData(): Uint8Array;
+    /**
+     * Get time-domain data from analyser for waveform visualization
+     */
+    getWaveformData(): Uint8Array;
     /**
-     * High-precision method to get visemes that should be active
-     * at the current playback frame. Use this in a requestAnimationFrame loop.
+     * Cleanup and close AudioContext
      */
-    getFrameVisemes(): Viseme[];
+    cleanup(): void;
     /**
-     * Change the system prompt mid-conversation.
+     * Get current audio context state
      */
-    updatePrompt(newPrompt: string): void;
+    getState(): 'running' | 'suspended' | 'closed' | 'interrupted' | null;
     /**
-     * Disconnect and release audio resources.
+     * Check if microphone is currently listening
      */
-    disconnect(): void;
+    isRecording(): boolean;
 }
 
-export { AUDIO_CONFIG, type AnalyserConfig, type BrowserAudioConfig, BrowserAudioManager, type BrowserAudioOptions, DEFAULT_URLS, Language, type LokutorConfig, StreamResampler, type SynthesizeOptions, TTSClient, type Viseme, VoiceAgent, VoiceAgentClient, type VoiceAgentOptions, VoiceStyle, applyLowPassFilter, bytesToPcm16, calculateRMS, float32ToPcm16, normalizeAudio, pcm16ToBytes, pcm16ToFloat32, resample, resampleWithAntiAliasing, simpleConversation, simpleTTS };
+export { AUDIO_CONFIG, type AnalyserConfig, type BrowserAudioConfig, BrowserAudioManager, type BrowserAudioOptions, DEFAULT_URLS, Language, type LokutorConfig, StreamResampler, type SynthesizeOptions, TTSClient, type Viseme, VoiceAgentClient, type VoiceAgentOptions, VoiceStyle, applyLowPassFilter, bytesToPcm16, calculateRMS, float32ToPcm16, normalizeAudio, pcm16ToBytes, pcm16ToFloat32, resample, resampleWithAntiAliasing, simpleConversation, simpleTTS };