sherpa-onnx-node 1.13.1 → 1.13.3

package/keyword-spotter.js

@@ -1,4 +1,5 @@
 /** @typedef {import('./types').KeywordSpotterConfig} KeywordSpotterConfig */
+/** @typedef {import('./types').KeywordSpotterHandle} KeywordSpotterHandle */
 /** @typedef {import('./types').KeywordResult} KeywordResult */
 /** @typedef {import('./streaming-asr').OnlineStream} OnlineStream */
 

package/non-streaming-asr.js

@@ -1,4 +1,6 @@
 /** @typedef {import('./types').OfflineStreamObject} OfflineStreamObject */
+/** @typedef {import('./types').OfflineStreamHandle} OfflineStreamHandle */
+/** @typedef {import('./types').OfflineRecognizerHandle} OfflineRecognizerHandle */
 /** @typedef {import('./types').Waveform} Waveform */
 /**
  * @typedef {import('./types').OfflineRecognizerConfig} OfflineRecognizerConfig

package/non-streaming-speaker-diarization.js

@@ -1,4 +1,5 @@
 /** @typedef {import('./types').OfflineSpeakerDiarizationConfig} OfflineSpeakerDiarizationConfig */
+/** @typedef {import('./types').OfflineSpeakerDiarizationHandle} OfflineSpeakerDiarizationHandle */
 /** @typedef {import('./types').SpeakerDiarizationSegment} SpeakerDiarizationSegment */
 
 const addon = require('./addon.js');

package/non-streaming-speech-denoiser.js

@@ -1,4 +1,5 @@
 /** @typedef {import('./types').OfflineSpeechDenoiserConfig} OfflineSpeechDenoiserConfig */
+/** @typedef {import('./types').OfflineSpeechDenoiserHandle} OfflineSpeechDenoiserHandle */
 /** @typedef {import('./types').GeneratedAudio} GeneratedAudio */
 /** @typedef {import('./types').AudioProcessRequest} AudioProcessRequest */
 

package/non-streaming-tts.js

@@ -1,4 +1,5 @@
 /** @typedef {import('./types').OfflineTtsConfig} OfflineTtsConfig */
+/** @typedef {import('./types').OfflineTtsHandle} OfflineTtsHandle */
 /** @typedef {import('./types').TtsRequest} TtsRequest */
 /** @typedef {import('./types').GeneratedAudio} GeneratedAudio */
 
@@ -83,7 +84,7 @@ class OfflineTts {
    *
    * The progress callback receives streaming audio chunks.
    *
-   * @param {TtsRequest & { generationConfig?: object, onProgress?: (info: {
+   * @param {TtsRequest & { generationConfig?: GenerationConfig, onProgress?: (info: {
    *     samples: Float32Array, progress: number }) => number | boolean | void
    *     }} obj
    * @returns {Promise<GeneratedAudio>}

package/online-speech-denoiser.js

@@ -1,4 +1,5 @@
 /** @typedef {import('./types').OnlineSpeechDenoiserConfig} OnlineSpeechDenoiserConfig */
+/** @typedef {import('./types').OnlineSpeechDenoiserHandle} OnlineSpeechDenoiserHandle */
 /** @typedef {import('./types').GeneratedAudio} GeneratedAudio */
 /** @typedef {import('./types').AudioProcessRequest} AudioProcessRequest */
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "sherpa-onnx-node",
-  "version": "1.13.1",
+  "version": "1.13.3",
   "description": "Speech-to-text, text-to-speech, speaker diarization, and speech enhancement using Next-gen Kaldi without internet connection",
   "main": "sherpa-onnx.js",
   "scripts": {
@@ -57,11 +57,11 @@
   },
   "homepage": "https://github.com/csukuangfj/sherpa-onnx#readme",
   "optionalDependencies": {
-    "sherpa-onnx-darwin-arm64": "^1.13.1",
-    "sherpa-onnx-darwin-x64": "^1.13.1",
-    "sherpa-onnx-linux-x64": "^1.13.1",
-    "sherpa-onnx-linux-arm64": "^1.13.1",
-    "sherpa-onnx-win-x64": "^1.13.1",
-    "sherpa-onnx-win-ia32": "^1.13.1"
+    "sherpa-onnx-darwin-arm64": "^1.13.3",
+    "sherpa-onnx-darwin-x64": "^1.13.3",
+    "sherpa-onnx-linux-x64": "^1.13.3",
+    "sherpa-onnx-linux-arm64": "^1.13.3",
+    "sherpa-onnx-win-x64": "^1.13.3",
+    "sherpa-onnx-win-ia32": "^1.13.3"
   }
 }

package/punctuation.js

@@ -1,6 +1,7 @@
 /** @typedef {import('./types').OfflinePunctuationHandle} OfflinePunctuationHandle */
 /** @typedef {import('./types').OfflinePunctuationConfig} OfflinePunctuationConfig */
 /** @typedef {import('./types').OnlinePunctuationConfig} OnlinePunctuationConfig */
+/** @typedef {import('./types').OnlinePunctuationHandle} OnlinePunctuationHandle */
 
 const addon = require('./addon.js');
 

package/resampler.js

@@ -0,0 +1,80 @@
+/** @typedef {import('./types').LinearResamplerHandle} LinearResamplerHandle */
+
+const addon = require('./addon.js');
+
+/**
+ * A linear resampler that converts audio from one sample rate to another.
+ */
+class LinearResampler {
+  /**
+   * Create a linear resampler.
+   *
+   * @param {number} inputSampleRate - Input sample rate in Hz.
+   * @param {number} outputSampleRate - Output sample rate in Hz.
+   */
+  constructor(inputSampleRate, outputSampleRate) {
+    /** @type {LinearResamplerHandle} */
+    this.handle =
+        addon.createLinearResampler(inputSampleRate, outputSampleRate);
+    this.inputSampleRate = inputSampleRate;
+    this.outputSampleRate = outputSampleRate;
+  }
+
+  /**
+   * Resample a chunk of audio samples.
+   *
+   * Call this for each chunk of input audio. For the final chunk, call
+   * {@link flush} instead so that any internally buffered samples are
+   * emitted.
+   *
+   * @param {Float32Array} samples - Input audio samples.
+   * @returns {Float32Array} Resampled audio samples.
+   */
+  resample(samples) {
+    return addon.resampleLinear(this.handle, samples, 0);
+  }
+
+  /**
+   * Resample the final chunk of audio and flush internal buffers.
+   *
+   * This is the same as {@link resample} but sets flush=1 so that any
+   * remaining samples buffered inside the resampler are emitted. Call
+   * this once after the last chunk of input audio.
+   *
+   * @param {Float32Array} samples - The final chunk of input audio samples.
+   * @returns {Float32Array} Resampled audio samples including buffered tail.
+   */
+  flush(samples) {
+    return addon.resampleLinear(this.handle, samples, 1);
+  }
+
+  /**
+   * Reset the resampler to its initial state, discarding any internal
+   * buffered samples.
+   */
+  reset() {
+    addon.linearResamplerReset(this.handle);
+  }
+
+  /**
+   * Get the input sample rate.
+   *
+   * @returns {number} Input sample rate in Hz.
+   */
+  getInputSampleRate() {
+    return addon.linearResamplerGetInputSampleRate(this.handle);
+  }
+
+  /**
+   * Get the output sample rate.
+   *
+   * @returns {number} Output sample rate in Hz.
+   */
+  getOutputSampleRate() {
+    return addon.linearResamplerGetOutputSampleRate(this.handle);
+  }
+}
+
+module.exports = {
+  LinearResampler,
+}

package/sherpa-onnx.js

@@ -19,6 +19,7 @@ const kws = require('./keyword-spotter.js');
 const sd = require('./non-streaming-speaker-diarization.js');
 const speech_denoiser = require('./non-streaming-speech-denoiser.js');
 const online_speech_denoiser = require('./online-speech-denoiser.js');
+const resampler = require('./resampler.js');
 
 module.exports = {
   OnlineRecognizer : streaming_asr.OnlineRecognizer,
@@ -40,6 +41,7 @@ module.exports = {
   OfflineSpeakerDiarization : sd.OfflineSpeakerDiarization,
   OfflineSpeechDenoiser : speech_denoiser.OfflineSpeechDenoiser,
   OnlineSpeechDenoiser : online_speech_denoiser.OnlineSpeechDenoiser,
+  LinearResampler : resampler.LinearResampler,
   version : addon.version,
   gitSha1 : addon.gitSha1,
   gitDate : addon.gitDate,

package/speaker-identification.js

@@ -2,6 +2,8 @@
 /** @typedef {import('./types').SpeakerEmbeddingManagerSearchObj} SpeakerEmbeddingManagerSearchObj */
 /** @typedef {import('./types').SpeakerEmbeddingManagerVerifyObj} SpeakerEmbeddingManagerVerifyObj */
 /** @typedef {import('./types').SpeakerEmbeddingExtractorConfig} SpeakerEmbeddingExtractorConfig */
+/** @typedef {import('./types').SpeakerEmbeddingExtractorHandle} SpeakerEmbeddingExtractorHandle */
+/** @typedef {import('./types').SpeakerEmbeddingManagerHandle} SpeakerEmbeddingManagerHandle */
 /** @typedef {import('./streaming-asr').OnlineStream} OnlineStream */
 
 const addon = require('./addon.js');

package/spoken-language-identification.js

@@ -1,4 +1,5 @@
 /** @typedef {import('./types').SpokenLanguageIdentificationConfig} SpokenLanguageIdentificationConfig */
+/** @typedef {import('./types').SpokenLanguageIdentificationHandle} SpokenLanguageIdentificationHandle */
 /** @typedef {import('./non-streaming-asr').OfflineStream} OfflineStream */
 
 const addon = require('./addon.js');

package/streaming-asr.js

@@ -1,5 +1,7 @@
 /** @typedef {import('./types').OnlineStreamObject} OnlineStreamObject */
 /** @typedef {import('./types').OnlineRecognizerHandle} OnlineRecognizerHandle */
+/** @typedef {import('./types').OnlineStreamHandle} OnlineStreamHandle */
+/** @typedef {import('./types').DisplayHandle} DisplayHandle */
 /** @typedef {import('./types').DisplayObject} DisplayObject */
 /** @typedef {import('./types').OnlineRecognizerConfig} OnlineRecognizerConfig */
 /** @typedef {import('./types').Waveform} Waveform */

package/types.js

@@ -52,7 +52,57 @@
 
 /**
  * @typedef {Object} OfflinePunctuationHandle
- * @see src/offline-punctuation.cc
+ * @see src/offline-punctuation.cc
+ */
+
+/**
+ * @typedef {Object} LinearResamplerHandle
+ * @see src/resampler.cc
+ */
+
+/**
+ * @typedef {Object} OfflineTtsHandle
+ * @see src/non-streaming-tts.cc
+ */
+
+/**
+ * @typedef {Object} OnlinePunctuationHandle
+ * @see src/punctuation.cc
+ */
+
+/**
+ * @typedef {Object} KeywordSpotterHandle
+ * @see src/keyword-spotter.cc
+ */
+
+/**
+ * @typedef {Object} SpeakerEmbeddingExtractorHandle
+ * @see src/speaker-identification.cc
+ */
+
+/**
+ * @typedef {Object} SpeakerEmbeddingManagerHandle
+ * @see src/speaker-identification.cc
+ */
+
+/**
+ * @typedef {Object} SpokenLanguageIdentificationHandle
+ * @see src/spoken-language-identification.cc
+ */
+
+/**
+ * @typedef {Object} OfflineSpeakerDiarizationHandle
+ * @see src/non-streaming-speaker-diarization.cc
+ */
+
+/**
+ * @typedef {Object} OfflineSpeechDenoiserHandle
+ * @see src/non-streaming-speech-denoiser.cc
+ */
+
+/**
+ * @typedef {Object} OnlineSpeechDenoiserHandle
+ * @see src/online-speech-denoiser.cc
  */
 
 /**
@@ -234,19 +284,19 @@
  * @property {number} [useInverseTextNormalization]
  */
 
-/**
- * Offline Cohere Transcribe model config
- * @typedef {Object} OfflineCohereTranscribeModelConfig
- * @property {string} [encoder]
- * @property {string} [decoder]
- * @property {string} [language]
- * @property {number} [usePunct]
- * @property {number} [useItn]
- */
-
-/**
- * Offline model config.
- * @typedef {Object} OfflineModelConfig
+/**
+ * Offline Cohere Transcribe model config
+ * @typedef {Object} OfflineCohereTranscribeModelConfig
+ * @property {string} [encoder]
+ * @property {string} [decoder]
+ * @property {string} [language]
+ * @property {number} [usePunct]
+ * @property {number} [useItn]
+ */
+
+/**
+ * Offline model config.
+ * @typedef {Object} OfflineModelConfig
  * @property {OfflineTransducerModelConfig} [transducer]
  * @property {OfflineParaformerModelConfig} [paraformer]
  * @property {OfflineZipformerCtcModelConfig} [zipformerCtc]
@@ -255,14 +305,14 @@
  * @property {OfflineMedAsrCtcModelConfig} [medasr]
  * @property {OfflineDolphinModelConfig} [dolphin]
  * @property {OfflineNeMoCtcModelConfig} [nemoCtc]
- * @property {OfflineCanaryModelConfig} [canary]
- * @property {OfflineWhisperModelConfig} [whisper]
- * @property {OfflineFireRedAsrModelConfig} [fireRedAsr]
-  * @property {OfflineMoonshineModelConfig} [moonshine]
-  * @property {OfflineTdnnModelConfig} [tdnn]
-  * @property {OfflineSenseVoiceModelConfig} [senseVoice]
- * @property {OfflineCohereTranscribeModelConfig} [cohereTranscribe]
- * @property {string} [tokens]
+ * @property {OfflineCanaryModelConfig} [canary]
+ * @property {OfflineWhisperModelConfig} [whisper]
+ * @property {OfflineFireRedAsrModelConfig} [fireRedAsr]
+ * @property {OfflineMoonshineModelConfig} [moonshine]
+ * @property {OfflineTdnnModelConfig} [tdnn]
+ * @property {OfflineSenseVoiceModelConfig} [senseVoice]
+ * @property {OfflineCohereTranscribeModelConfig} [cohereTranscribe]
+ * @property {string} [tokens]
  * @property {number} [numThreads]
  * @property {boolean|number} [debug]
  * @property {string} [provider]
@@ -512,24 +562,24 @@
  * @property {number} [lengthScale]
  */
 
-/**
- * @typedef {Object} OfflineTtsZipvoiceModelConfig
- * @property {string} [tokens]
- * @property {string} [encoder]
- * @property {string} [decoder]
- * @property {string} [vocoder]
- * @property {string} [dataDir]
- * @property {string} [lexicon]
- * @property {number} [featScale]
- * @property {number} [tShift]
- * @property {number} [targetRms]
- * @property {number} [guidanceScale]
- */
-
-/**
- * @typedef {Object} OfflineTtsPocketModelConfig
- * @property {string} [lmFlow]
- * @property {string} [lmMain]
+/**
+ * @typedef {Object} OfflineTtsZipvoiceModelConfig
+ * @property {string} [tokens]
+ * @property {string} [encoder]
+ * @property {string} [decoder]
+ * @property {string} [vocoder]
+ * @property {string} [dataDir]
+ * @property {string} [lexicon]
+ * @property {number} [featScale]
+ * @property {number} [tShift]
+ * @property {number} [targetRms]
+ * @property {number} [guidanceScale]
+ */
+
+/**
+ * @typedef {Object} OfflineTtsPocketModelConfig
+ * @property {string} [lmFlow]
+ * @property {string} [lmMain]
  * @property {string} [encoder]
  * @property {string} [decoder]
  * @property {string} [textConditioner]
@@ -542,12 +592,12 @@
  * Offline TTS model config
  * @typedef {Object} OfflineTtsModelConfig
  * @property {OfflineTtsVitsModelConfig} [vits]
- * @property {OfflineTtsMatchaModelConfig} [matcha]
- * @property {OfflineTtsKokoroModelConfig} [kokoro]
- * @property {OfflineTtsKittenModelConfig} [kitten]
- * @property {OfflineTtsZipvoiceModelConfig} [zipvoice]
- * @property {OfflineTtsPocketModelConfig} [pocket]
- */
+ * @property {OfflineTtsMatchaModelConfig} [matcha]
+ * @property {OfflineTtsKokoroModelConfig} [kokoro]
+ * @property {OfflineTtsKittenModelConfig} [kitten]
+ * @property {OfflineTtsZipvoiceModelConfig} [zipvoice]
+ * @property {OfflineTtsPocketModelConfig} [pocket]
+ */
 
 /**
  * Offline TTS configuration (partial, commonly used props).
@@ -559,39 +609,39 @@
  * @property {string} [provider]
  */
 
-/**
- * Offline Speech Denoiser model config
- * @typedef {Object} OfflineSpeechDenoiserGtcrnModelConfig
- * @property {string} [model]
- */
-
-/**
- * Offline Speech Denoiser model config
- * @typedef {Object} OfflineSpeechDenoiserDpdfNetModelConfig
- * @property {string} [model]
- */
-
-/**
- * Offline Speech Denoiser model config
- * @typedef {Object} OfflineSpeechDenoiserModelConfig
- * @property {OfflineSpeechDenoiserGtcrnModelConfig} [gtcrn]
- * @property {OfflineSpeechDenoiserDpdfNetModelConfig} [dpdfnet]
- * @property {number} [numThreads]
- * @property {boolean|number} [debug]
- * @property {string} [provider]
- */
-
-/**
- * Offline Speech Denoiser configuration (partial).
- * @typedef {Object} OfflineSpeechDenoiserConfig
- * @property {OfflineSpeechDenoiserModelConfig} [model]
- */
-
-/**
- * Online Speech Denoiser configuration (partial).
- * @typedef {Object} OnlineSpeechDenoiserConfig
- * @property {OfflineSpeechDenoiserModelConfig} [model]
- */
+/**
+ * Offline Speech Denoiser model config
+ * @typedef {Object} OfflineSpeechDenoiserGtcrnModelConfig
+ * @property {string} [model]
+ */
+
+/**
+ * Offline Speech Denoiser model config
+ * @typedef {Object} OfflineSpeechDenoiserDpdfNetModelConfig
+ * @property {string} [model]
+ */
+
+/**
+ * Offline Speech Denoiser model config
+ * @typedef {Object} OfflineSpeechDenoiserModelConfig
+ * @property {OfflineSpeechDenoiserGtcrnModelConfig} [gtcrn]
+ * @property {OfflineSpeechDenoiserDpdfNetModelConfig} [dpdfnet]
+ * @property {number} [numThreads]
+ * @property {boolean|number} [debug]
+ * @property {string} [provider]
+ */
+
+/**
+ * Offline Speech Denoiser configuration (partial).
+ * @typedef {Object} OfflineSpeechDenoiserConfig
+ * @property {OfflineSpeechDenoiserModelConfig} [model]
+ */
+
+/**
+ * Online Speech Denoiser configuration (partial).
+ * @typedef {Object} OnlineSpeechDenoiserConfig
+ * @property {OfflineSpeechDenoiserModelConfig} [model]
+ */
 
 /**
  * Offline speaker segmentation (pyannote) model config

package/vad.js

@@ -1,4 +1,5 @@
 /** @typedef {import('./types').CircularBufferHandle} CircularBufferHandle */
+/** @typedef {import('./types').VoiceActivityDetectorHandle} VoiceActivityDetectorHandle */
 /** @typedef {import('./types').SpeechSegment} SpeechSegment */
 /** @typedef {import('./types').VadConfig} VadConfig */