@happyvertical/smrt-voice 0.51.31 → 0.51.33

package/dist/__smrt-register__.d.ts

@@ -0,0 +1,2 @@
+export {};
+//# sourceMappingURL=__smrt-register__.d.ts.map

package/dist/__smrt-register__.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"__smrt-register__.d.ts","sourceRoot":"","sources":["../src/__smrt-register__.ts"],"names":[],"mappings":""}

package/dist/index.d.ts

@@ -1,587 +1,40 @@
-import { AudioInput } from '@happyvertical/speech';
-import { Content } from '@happyvertical/smrt-content';
-import { ContentOptions } from '@happyvertical/smrt-content';
-import { SmrtObject } from '@happyvertical/smrt-core';
-import { SmrtObjectOptions } from '@happyvertical/smrt-core';
-import { Speech } from '@happyvertical/speech';
-import { SpeechAdapterAvailability } from '@happyvertical/speech';
-import { SpeechSynthesizer } from '@happyvertical/speech';
-import { SpeechSynthesizerType } from '@happyvertical/speech';
-import { SpeechVoice } from '@happyvertical/speech';
-import { SpeechVoiceInput } from '@happyvertical/speech';
-import { WordTiming as SpeechWordTiming } from '@happyvertical/speech';
-import { SynthesisRequest } from '@happyvertical/speech';
-import { SynthesizedSpeech } from '@happyvertical/speech';
-import { Transcriber } from '@happyvertical/speech';
-import { TranscriberType } from '@happyvertical/speech';
-import { TranscriptionRequest } from '@happyvertical/speech';
-import { TranscriptResult } from '@happyvertical/speech';
-import { TranscriptSegment } from '@happyvertical/speech';
-
-export { AudioInput }
-
-export declare function metadataFromSynthesizedSpeech(speech: SynthesizedSpeech): VoiceOutputMetadata;
-
-/**
- * Audio sample quality rating
- */
-export declare type SampleQuality = 'low' | 'medium' | 'high';
-
-export { Speech }
-
-export { SpeechAdapterAvailability }
-
-export { SpeechSynthesizer }
-
-export { SpeechSynthesizerType }
-
-export { SpeechVoice }
-
-export { SpeechVoiceInput }
-
-export { SpeechWordTiming }
-
-export { SynthesisRequest }
-
-export { SynthesizedSpeech }
-
-export { Transcriber }
-
-export { TranscriberType }
-
-export { TranscriptionRequest }
-
-export { TranscriptResult }
-
-export { TranscriptSegment }
-
-/**
- * Voice gender classification
- */
-export declare type VoiceGender = 'male' | 'female' | 'neutral';
-
-/**
- * Generated audio output from text-to-speech synthesis
- *
- * VoiceOutput extends Content to represent audio generated from
- * text using a VoiceProfile. It includes word-level timing information
- * for lip-sync alignment in video production.
- *
- * @example
- * ```typescript
- * import { VoiceOutput } from '@happyvertical/smrt-voice';
- *
- * const output = new VoiceOutput({
- *   voiceProfileId: 'voice-123',
- *   sourceText: 'Welcome to the evening news broadcast.',
- *   audioAssetId: 'asset-789',
- *   duration: 3.5,
- *   wordTimings: [
- *     { word: 'Welcome', start: 0.0, end: 0.4 },
- *     { word: 'to', start: 0.4, end: 0.5 },
- *     { word: 'the', start: 0.5, end: 0.6 },
- *     { word: 'evening', start: 0.6, end: 1.0 },
- *     { word: 'news', start: 1.0, end: 1.3 },
- *     { word: 'broadcast', start: 1.3, end: 1.9 },
- *   ],
- *   audioMetadata: {
- *     sampleRate: 48000,
- *     format: 'wav',
- *     channels: 1,
- *     provider: 'qwen3-tts',
- *   },
- * });
- * ```
- */
-export declare class VoiceOutput extends Content {
-    /**
-     * Voice profile used for synthesis
-     */
-    voiceProfileId: string | null;
-    /**
-     * Original text that was synthesized
-     */
-    sourceText: string;
-    /**
-     * Asset ID of the generated audio file
-     */
-    audioAssetId: string | null;
-    /**
-     * Duration of the generated audio in seconds
-     */
-    duration: number;
-    /**
-     * Word-level timing information for lip-sync alignment
-     */
-    wordTimings: WordTiming[] | null;
-    /**
-     * Audio metadata (sample rate, format, etc.)
-     */
-    audioMetadata: VoiceOutputMetadata;
-    constructor(options?: VoiceOutputOptions);
-    /**
-     * Get the word count of the source text
-     */
-    get wordCount(): number;
-    /**
-     * Get the average words per second rate
-     */
-    get wordsPerSecond(): number;
-    /**
-     * Check if word timing data is available for lip-sync
-     */
-    get hasWordTimings(): boolean;
-    /**
-     * Get the word at a specific timestamp
-     */
-    getWordAtTime(seconds: number): WordTiming | null;
-    /**
-     * Build a persisted output model from a provider-neutral speech result.
-     *
-     * The binary audio remains in the caller-owned asset pipeline; this helper
-     * only normalizes metadata and timing fields onto the SMRT model.
-     */
-    static fromSynthesizedSpeech(options: VoiceOutputFromSynthesizedSpeechOptions): VoiceOutput;
-}
-
-/**
- * Create a VoiceOutput from a runtime @happyvertical/speech synthesis result.
- */
-export declare interface VoiceOutputFromSynthesizedSpeechOptions extends VoiceOutputOptions {
-    /**
-     * Runtime synthesis result from @happyvertical/speech.
-     */
-    speech: SynthesizedSpeech;
-}
-
-/**
- * Voice output metadata
- */
-export declare interface VoiceOutputMetadata {
-    /**
-     * Sample rate in Hz
-     */
-    sampleRate?: number;
-    /**
-     * Audio format (e.g., 'wav', 'mp3', 'ogg')
-     */
-    format?: string;
-    /**
-     * MIME type returned by the speech provider
-     */
-    contentType?: string;
-    /**
-     * Number of audio channels
-     */
-    channels?: number;
-    /**
-     * Bit depth (e.g., 16, 24, 32)
-     */
-    bitDepth?: number;
-    /**
-     * File size in bytes
-     */
-    fileSize?: number;
-    /**
-     * TTS provider used
-     */
-    provider?: string;
-    /**
-     * Model used for synthesis
-     */
-    model?: string;
-    /**
-     * Speech speed used (1.0 = normal)
-     */
-    speed?: number;
-    /**
-     * Pitch adjustment used (semitones)
-     */
-    pitch?: number;
-}
-
-/**
- * Voice output creation options
- */
-export declare interface VoiceOutputOptions extends ContentOptions {
-    /**
-     * Voice profile used for synthesis
-     */
-    voiceProfileId?: string | null;
-    /**
-     * Original text that was synthesized
-     */
-    sourceText?: string;
-    /**
-     * Asset ID of the generated audio file
-     */
-    audioAssetId?: string | null;
-    /**
-     * Duration of the generated audio in seconds
-     */
-    duration?: number;
-    /**
-     * Word-level timing information for lip-sync
-     */
-    wordTimings?: WordTiming[] | null;
-    /**
-     * Audio metadata
-     */
-    audioMetadata?: VoiceOutputMetadata;
-}
-
-/**
- * Voice profile for AI-powered speech synthesis
- *
- * VoiceProfile represents a configured voice identity that can be used
- * for text-to-speech synthesis. Voices can be created through:
- * - Voice design: Natural language description of desired voice characteristics
- * - Voice cloning: 3+ second audio sample for voice replication
- *
- * @example
- * ```typescript
- * import { VoiceProfile } from '@happyvertical/smrt-voice';
- *
- * // Create a designed voice
- * const anchorVoice = new VoiceProfile({
- *   name: 'News Anchor',
- *   description: 'Professional news anchor voice with clear enunciation',
- *   language: 'en-US',
- *   gender: 'male',
- *   designPrompt: 'Warm, authoritative male voice with slight gravitas, suitable for news broadcasts',
- *   provider: 'qwen3-tts',
- * });
- *
- * // Create a cloned voice
- * const clonedVoice = new VoiceProfile({
- *   name: 'Custom Voice',
- *   description: 'Cloned from user sample',
- *   language: 'en-US',
- *   sampleAssetId: 'asset-123',
- *   provider: 'qwen3-tts',
- * });
- * ```
- */
-export declare class VoiceProfile extends SmrtObject {
-    /**
-     * Tenant ID for multi-tenant isolation
-     * Nullable to support global/default voices
-     */
-    tenantId: string | null;
-    /**
-     * Human-readable name for the voice profile
-     */
-    name: string;
-    /**
-     * Description of the voice characteristics
-     */
-    description: string | null;
-    /**
-     * ISO language code (e.g., 'en-US', 'zh-CN')
-     */
-    language: string;
-    /**
-     * Voice gender classification
-     */
-    gender: VoiceGender;
-    /**
-     * Natural language description for voice design
-     * Used when creating a voice from scratch via AI
-     */
-    designPrompt: string | null;
-    /**
-     * Asset ID of the audio sample for voice cloning
-     * Should be at least 3 seconds of clear speech
-     */
-    sampleAssetId: string | null;
-    /**
-     * Provider-specific voice data (ID, embedding, etc.)
-     * Stored after voice creation/cloning is complete
-     */
-    voiceData: Record<string, unknown> | null;
-    /**
-     * Default speech speed multiplier (0.5 - 2.0)
-     * 1.0 = normal speed
-     */
-    defaultSpeed: number;
-    /**
-     * Default pitch adjustment in semitones (-20 to 20)
-     * 0 = no adjustment
-     */
-    defaultPitch: number;
-    /**
-     * Voice profile status
-     * - pending: Profile created but voice not yet generated
-     * - processing: Voice generation/cloning in progress
-     * - ready: Voice is ready for use
-     * - failed: Voice generation failed
-     */
-    status: VoiceProfileStatus;
-    /**
-     * TTS provider that created/manages this voice
-     */
-    provider: string;
-    /**
-     * Error message if status is 'failed'
-     */
-    errorMessage: string | null;
-    constructor(options?: VoiceProfileOptions);
-    /**
-     * Check if this voice profile uses voice cloning
-     */
-    get isCloned(): boolean;
-    /**
-     * Check if this voice profile uses voice design
-     */
-    get isDesigned(): boolean;
-    /**
-     * Check if the voice is ready for use
-     */
-    get isReady(): boolean;
-    /**
-     * Check if this is a global (default) voice
-     */
-    get isGlobal(): boolean;
-    /**
-     * Convert this persisted profile into the runtime voice shape consumed by
-     * @happyvertical/speech adapters.
-     */
-    toSpeechVoice(): SpeechVoice;
-}
-
-/**
- * Voice profile creation options
- */
-export declare interface VoiceProfileOptions extends SmrtObjectOptions {
-    /**
-     * Human-readable name for the voice profile
-     */
-    name?: string;
-    /**
-     * Description of the voice characteristics
-     */
-    description?: string | null;
-    /**
-     * ISO language code (e.g., 'en-US', 'zh-CN')
-     */
-    language?: string;
-    /**
-     * Voice gender classification
-     * @default 'neutral'
-     */
-    gender?: VoiceGender;
-    /**
-     * Natural language description for voice design
-     * Used when creating a voice from scratch
-     */
-    designPrompt?: string | null;
-    /**
-     * Asset ID of the audio sample for voice cloning
-     * Should be at least 3 seconds of clear speech
-     */
-    sampleAssetId?: string | null;
-    /**
-     * Provider-specific voice data (ID, embedding, etc.)
-     * Stored after voice creation/cloning
-     */
-    voiceData?: Record<string, unknown> | null;
-    /**
-     * Default speech speed multiplier
-     * @default 1.0
-     */
-    defaultSpeed?: number;
-    /**
-     * Default pitch adjustment in semitones
-     * @default 0
-     */
-    defaultPitch?: number;
-    /**
-     * Voice profile status
-     * @default 'pending'
-     */
-    status?: VoiceProfileStatus;
-    /**
-     * TTS provider that created this voice
-     * @default 'qwen3-tts'
-     */
-    provider?: string;
-    /**
-     * Error message if status is 'failed'
-     */
-    errorMessage?: string | null;
-    /**
-     * Tenant ID for multi-tenant isolation
-     * Null for global/default voices
-     */
-    tenantId?: string | null;
-}
-
-/**
- * Voice profile status
- */
-export declare type VoiceProfileStatus = 'pending' | 'processing' | 'ready' | 'failed';
-
-/**
- * Audio sample for voice cloning
- *
- * VoiceSample represents an audio recording used as source material
- * for voice cloning. For best results, samples should be:
- * - At least 3 seconds long
- * - Clear speech without background noise
- * - Single speaker only
- * - High quality (44.1kHz or higher)
- *
- * Multiple samples can be associated with a single VoiceProfile
- * to improve voice cloning quality.
- *
- * @example
- * ```typescript
- * import { VoiceSample } from '@happyvertical/smrt-voice';
- *
- * const sample = new VoiceSample({
- *   voiceProfileId: 'voice-123',
- *   assetId: 'asset-456',
- *   duration: 5.2,
- *   transcription: 'Hello, this is a test recording for voice cloning.',
- *   quality: 'high',
- *   sampleRate: 48000,
- *   channels: 1,
- *   format: 'wav',
- *   isPrimary: true,
- * });
- * ```
- */
-export declare class VoiceSample extends SmrtObject {
-    /**
-     * Tenant ID for multi-tenant isolation
-     */
-    tenantId: string | null;
-    /**
-     * Voice profile this sample belongs to
-     */
-    voiceProfileId: string | null;
-    /**
-     * Asset ID of the audio file
-     * References an Asset in smrt-assets
-     */
-    assetId: string | null;
-    /**
-     * Sample duration in seconds
-     */
-    duration: number;
-    /**
-     * Transcription of what was said in the sample
-     * Used for alignment and quality verification
-     */
-    transcription: string | null;
-    /**
-     * Quality rating based on audio analysis
-     * - low: Noisy or short samples
-     * - medium: Acceptable quality
-     * - high: Clear audio, good length
-     */
-    quality: SampleQuality;
-    /**
-     * Sample rate in Hz
-     */
-    sampleRate: number | null;
-    /**
-     * Number of audio channels
-     */
-    channels: number | null;
-    /**
-     * Audio format
-     */
-    format: string | null;
-    /**
-     * Whether this is the primary sample for the voice profile
-     */
-    isPrimary: boolean;
-    constructor(options?: VoiceSampleOptions);
-    /**
-     * Check if sample meets minimum duration requirement (3 seconds)
-     */
-    get meetsMinDuration(): boolean;
-    /**
-     * Check if sample is high quality and suitable for cloning
-     */
-    get isSuitableForCloning(): boolean;
-}
-
-/**
- * Voice sample creation options
- */
-export declare interface VoiceSampleOptions extends SmrtObjectOptions {
-    /**
-     * Voice profile this sample belongs to
-     */
-    voiceProfileId?: string | null;
-    /**
-     * Asset ID of the audio file
-     */
-    assetId?: string | null;
-    /**
-     * Sample duration in seconds
-     */
-    duration?: number;
-    /**
-     * Transcription of what was said in the sample
-     */
-    transcription?: string | null;
-    /**
-     * Quality rating based on audio analysis
-     * @default 'medium'
-     */
-    quality?: SampleQuality;
-    /**
-     * Sample rate in Hz (e.g., 44100, 48000)
-     */
-    sampleRate?: number | null;
-    /**
-     * Number of audio channels (1 = mono, 2 = stereo)
-     */
-    channels?: number | null;
-    /**
-     * Audio format (e.g., 'wav', 'mp3', 'ogg')
-     */
-    format?: string | null;
-    /**
-     * Whether this is the primary sample for the voice profile
-     * @default false
-     */
-    isPrimary?: boolean;
-    /**
-     * Tenant ID for multi-tenant isolation
-     */
-    tenantId?: string | null;
-}
-
-/**
- * Word timing information for lip-sync alignment
- */
-export declare interface WordTiming {
-    /**
-     * The word
-     */
-    word: string;
-    /**
-     * Start time in seconds
-     */
-    start: number;
-    /**
-     * End time in seconds
-     */
-    end: number;
-    /**
-     * Provider confidence score, when available
-     */
-    confidence?: number;
-    /**
-     * Speaker identifier, when diarization is available
-     */
-    speakerId?: string;
-}
-
-export declare function wordTimingsFromSpeech(words: readonly SpeechWordTiming[] | null | undefined): WordTiming[] | null;
-
-export { }
+/**
+ * @happyvertical/smrt-voice
+ *
+ * Voice profile management for AI-powered voice synthesis and cloning.
+ *
+ * This package provides models for managing voice profiles, audio samples,
+ * and synthesized voice outputs in the SMRT ecosystem.
+ *
+ * @example
+ * ```typescript
+ * import {
+ *   VoiceProfile,
+ *   VoiceSample,
+ *   VoiceOutput,
+ * } from '@happyvertical/smrt-voice';
+ *
+ * // Create a voice profile with voice design
+ * const profile = new VoiceProfile({
+ *   name: 'News Anchor',
+ *   language: 'en-US',
+ *   gender: 'male',
+ *   designPrompt: 'Warm, authoritative male voice with clear enunciation',
+ * });
+ * await profile.save();
+ *
+ * // Or create from a cloned voice sample
+ * const clonedProfile = new VoiceProfile({
+ *   name: 'Custom Voice',
+ *   language: 'en-US',
+ *   sampleAssetId: 'asset-123',
+ * });
+ * await clonedProfile.save();
+ * ```
+ */
+import './__smrt-register__.js';
+export type { AudioInput, Speech, SpeechAdapterAvailability, SpeechSynthesizer, SpeechSynthesizerType, SpeechVoice, SpeechVoiceInput, SynthesisRequest, SynthesizedSpeech, Transcriber, TranscriberType, TranscriptionRequest, TranscriptResult, TranscriptSegment, WordTiming as SpeechWordTiming, } from '@happyvertical/speech';
+export { metadataFromSynthesizedSpeech, VoiceOutput, type VoiceOutputFromSynthesizedSpeechOptions, type VoiceOutputMetadata, type VoiceOutputOptions, type WordTiming, wordTimingsFromSpeech, } from './voice-output.js';
+export { type VoiceGender, VoiceProfile, type VoiceProfileOptions, type VoiceProfileStatus, } from './voice-profile.js';
+export { type SampleQuality, VoiceSample, type VoiceSampleOptions, } from './voice-sample.js';
+//# sourceMappingURL=index.d.ts.map

package/dist/index.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../src/index.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GAiCG;AAKH,OAAO,wBAAwB,CAAC;AAEhC,YAAY,EACV,UAAU,EACV,MAAM,EACN,yBAAyB,EACzB,iBAAiB,EACjB,qBAAqB,EACrB,WAAW,EACX,gBAAgB,EAChB,gBAAgB,EAChB,iBAAiB,EACjB,WAAW,EACX,eAAe,EACf,oBAAoB,EACpB,gBAAgB,EAChB,iBAAiB,EACjB,UAAU,IAAI,gBAAgB,GAC/B,MAAM,uBAAuB,CAAC;AAC/B,OAAO,EACL,6BAA6B,EAC7B,WAAW,EACX,KAAK,uCAAuC,EAC5C,KAAK,mBAAmB,EACxB,KAAK,kBAAkB,EACvB,KAAK,UAAU,EACf,qBAAqB,GACtB,MAAM,mBAAmB,CAAC;AAE3B,OAAO,EACL,KAAK,WAAW,EAChB,YAAY,EACZ,KAAK,mBAAmB,EACxB,KAAK,kBAAkB,GACxB,MAAM,oBAAoB,CAAC;AAC5B,OAAO,EACL,KAAK,aAAa,EAClB,WAAW,EACX,KAAK,kBAAkB,GACxB,MAAM,mBAAmB,CAAC"}