Add JavaScript API (node-addon) for FireRedAsr (#1873)

@@ -10,6 +10,14 @@ arch=$(node -p "require('os').arch()")
 platform=$(node -p "require('os').platform()")
 node_version=$(node -p "process.versions.node.split('.')[0]")
 
+echo "----------non-streaming asr FireRedAsr----------"
+curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-fire-red-asr-large-zh_en-2025-02-16.tar.bz2
+tar xvf sherpa-onnx-fire-red-asr-large-zh_en-2025-02-16.tar.bz2
+rm sherpa-onnx-fire-red-asr-large-zh_en-2025-02-16.tar.bz2
+
+node ./test_asr_non_streaming_fire_red_asr.js
+rm -rf sherpa-onnx-fire-red-asr-large-zh_en-2025-02-16
+
 echo "----------non-streaming asr moonshine + vad----------"
 curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-moonshine-tiny-en-int8.tar.bz2
 tar xvf sherpa-onnx-moonshine-tiny-en-int8.tar.bz2

@@ -80,6 +80,23 @@ static SherpaOnnxOfflineWhisperModelConfig GetOfflineWhisperModelConfig(
   return c;
 }
 
+static SherpaOnnxOfflineFireRedAsrModelConfig GetOfflineFireRedAsrModelConfig(
+    Napi::Object obj) {
+  SherpaOnnxOfflineFireRedAsrModelConfig c;
+  memset(&c, 0, sizeof(c));
+
+  if (!obj.Has("fireRedAsr") || !obj.Get("fireRedAsr").IsObject()) {
+    return c;
+  }
+
+  Napi::Object o = obj.Get("fireRedAsr").As<Napi::Object>();
+
+  SHERPA_ONNX_ASSIGN_ATTR_STR(encoder, encoder);
+  SHERPA_ONNX_ASSIGN_ATTR_STR(decoder, decoder);
+
+  return c;
+}
+
 static SherpaOnnxOfflineMoonshineModelConfig GetOfflineMoonshineModelConfig(
     Napi::Object obj) {
   SherpaOnnxOfflineMoonshineModelConfig c;
@@ -150,6 +167,7 @@ static SherpaOnnxOfflineModelConfig GetOfflineModelConfig(Napi::Object obj) {
   c.tdnn = GetOfflineTdnnModelConfig(o);
   c.sense_voice = GetOfflineSenseVoiceModelConfig(o);
   c.moonshine = GetOfflineMoonshineModelConfig(o);
+  c.fire_red_asr = GetOfflineFireRedAsrModelConfig(o);
 
   SHERPA_ONNX_ASSIGN_ATTR_STR(tokens, tokens);
   SHERPA_ONNX_ASSIGN_ATTR_INT32(num_threads, numThreads);
@@ -271,6 +289,9 @@ CreateOfflineRecognizerWrapper(const Napi::CallbackInfo &info) {
   SHERPA_ONNX_DELETE_C_STR(c.model_config.moonshine.uncached_decoder);
   SHERPA_ONNX_DELETE_C_STR(c.model_config.moonshine.cached_decoder);
 
+  SHERPA_ONNX_DELETE_C_STR(c.model_config.fire_red_asr.encoder);
+  SHERPA_ONNX_DELETE_C_STR(c.model_config.fire_red_asr.decoder);
+
   SHERPA_ONNX_DELETE_C_STR(c.model_config.tokens);
   SHERPA_ONNX_DELETE_C_STR(c.model_config.provider);
   SHERPA_ONNX_DELETE_C_STR(c.model_config.model_type);

@@ -110,6 +110,7 @@ The following tables list the examples in this folder.
 |File| Description|
 |---|---|
 |[./test_asr_non_streaming_transducer.js](./test_asr_non_streaming_transducer.js)|Non-streaming speech recognition from a file with a Zipformer transducer model|
+|[./test_asr_non_streaming_fire_red_asr.js](./test_asr_non_streaming_fire_red_asr.js)| Non-streaming speech recognition from a file using [FireRedAsr](https://github.com/FireRedTeam/FireRedASR)|
 |[./test_asr_non_streaming_whisper.js](./test_asr_non_streaming_whisper.js)| Non-streaming speech recognition from a file using [Whisper](https://github.com/openai/whisper)|
 |[./test_vad_with_non_streaming_asr_whisper.js](./test_vad_with_non_streaming_asr_whisper.js)| Non-streaming speech recognition from a file using [Whisper](https://github.com/openai/whisper) + [Silero VAD](https://github.com/snakers4/silero-vad)|
 |[./test_asr_non_streaming_moonshine.js](./test_asr_non_streaming_moonshine.js)|Non-streaming speech recognition from a file using [Moonshine](https://github.com/usefulsensors/moonshine)|
@@ -253,6 +254,15 @@ npm install naudiodon2
 node ./test_vad_asr_non_streaming_transducer_microphone.js
 ```
 
+### Non-streaming speech recognition with FireRedAsr
+```bash
+wget https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-fire-red-asr-large-zh_en-2025-02-16.tar.bz2
+tar xvf sherpa-onnx-fire-red-asr-large-zh_en-2025-02-16.tar.bz2
+rm sherpa-onnx-fire-red-asr-large-zh_en-2025-02-16.tar.bz2
+
+node ./test_asr_non_streaming_fire_red_asr.js
+```
+
 ### Non-streaming speech recognition with Whisper
 
 ```bash

+// Copyright (c)  2025  Xiaomi Corporation
+const sherpa_onnx = require('sherpa-onnx-node');
+
+// Please download test files from
+// https://github.com/k2-fsa/sherpa-onnx/releases/tag/asr-models
+const config = {
+  'featConfig': {
+    'sampleRate': 16000,
+    'featureDim': 80,
+  },
+  'modelConfig': {
+    'fireRedAsr': {
+      'encoder':
+          './sherpa-onnx-fire-red-asr-large-zh_en-2025-02-16/encoder.int8.onnx',
+      'decoder':
+          './sherpa-onnx-fire-red-asr-large-zh_en-2025-02-16/decoder.int8.onnx',
+    },
+    'tokens': './sherpa-onnx-fire-red-asr-large-zh_en-2025-02-16/tokens.txt',
+    'numThreads': 2,
+    'provider': 'cpu',
+    'debug': 1,
+  }
+};
+
+const waveFilename =
+    './sherpa-onnx-fire-red-asr-large-zh_en-2025-02-16/test_wavs/0.wav';
+
+const recognizer = new sherpa_onnx.OfflineRecognizer(config);
+console.log('Started')
+let start = Date.now();
+const stream = recognizer.createStream();
+const wave = sherpa_onnx.readWave(waveFilename);
+stream.acceptWaveform({sampleRate: wave.sampleRate, samples: wave.samples});
+
+recognizer.decode(stream);
+result = recognizer.getResult(stream)
+let stop = Date.now();
+console.log('Done')
+
+const elapsed_seconds = (stop - start) / 1000;
+const duration = wave.samples.length / wave.sampleRate;
+const real_time_factor = elapsed_seconds / duration;
+console.log('Wave duration', duration.toFixed(3), 'seconds')
+console.log('Elapsed', elapsed_seconds.toFixed(3), 'seconds')
+console.log(
+    `RTF = ${elapsed_seconds.toFixed(3)}/${duration.toFixed(3)} =`,
+    real_time_factor.toFixed(3))
+console.log(waveFilename)
+console.log('result\n', result)

@@ -41,8 +41,8 @@ console.log('Done')
 const elapsed_seconds = (stop - start) / 1000;
 const duration = wave.samples.length / wave.sampleRate;
 const real_time_factor = elapsed_seconds / duration;
-console.log('Wave duration', duration.toFixed(3), 'secodns')
-console.log('Elapsed', elapsed_seconds.toFixed(3), 'secodns')
+console.log('Wave duration', duration.toFixed(3), 'seconds')
+console.log('Elapsed', elapsed_seconds.toFixed(3), 'seconds')
 console.log(
     `RTF = ${elapsed_seconds.toFixed(3)}/${duration.toFixed(3)} =`,
     real_time_factor.toFixed(3))

@@ -39,8 +39,8 @@ console.log('Done')
 const elapsed_seconds = (stop - start) / 1000;
 const duration = wave.samples.length / wave.sampleRate;
 const real_time_factor = elapsed_seconds / duration;
-console.log('Wave duration', duration.toFixed(3), 'secodns')
-console.log('Elapsed', elapsed_seconds.toFixed(3), 'secodns')
+console.log('Wave duration', duration.toFixed(3), 'seconds')
+console.log('Elapsed', elapsed_seconds.toFixed(3), 'seconds')
 console.log(
     `RTF = ${elapsed_seconds.toFixed(3)}/${duration.toFixed(3)} =`,
     real_time_factor.toFixed(3))

@@ -37,8 +37,8 @@ console.log('Done')
 const elapsed_seconds = (stop - start) / 1000;
 const duration = wave.samples.length / wave.sampleRate;
 const real_time_factor = elapsed_seconds / duration;
-console.log('Wave duration', duration.toFixed(3), 'secodns')
-console.log('Elapsed', elapsed_seconds.toFixed(3), 'secodns')
+console.log('Wave duration', duration.toFixed(3), 'seconds')
+console.log('Elapsed', elapsed_seconds.toFixed(3), 'seconds')
 console.log(
     `RTF = ${elapsed_seconds.toFixed(3)}/${duration.toFixed(3)} =`,
     real_time_factor.toFixed(3))

@@ -39,8 +39,8 @@ console.log('Done')
 const elapsed_seconds = (stop - start) / 1000;
 const duration = wave.samples.length / wave.sampleRate;
 const real_time_factor = elapsed_seconds / duration;
-console.log('Wave duration', duration.toFixed(3), 'secodns')
-console.log('Elapsed', elapsed_seconds.toFixed(3), 'secodns')
+console.log('Wave duration', duration.toFixed(3), 'seconds')
+console.log('Elapsed', elapsed_seconds.toFixed(3), 'seconds')
 console.log(
     `RTF = ${elapsed_seconds.toFixed(3)}/${duration.toFixed(3)} =`,
     real_time_factor.toFixed(3))

@@ -54,8 +54,8 @@ console.log('Done')
 const elapsed_seconds = (stop - start) / 1000;
 const duration = wave.samples.length / wave.sampleRate;
 const real_time_factor = elapsed_seconds / duration;
-console.log('Wave duration', duration.toFixed(3), 'secodns')
-console.log('Elapsed', elapsed_seconds.toFixed(3), 'secodns')
+console.log('Wave duration', duration.toFixed(3), 'seconds')
+console.log('Elapsed', elapsed_seconds.toFixed(3), 'seconds')
 console.log(
     `RTF = ${elapsed_seconds.toFixed(3)}/${duration.toFixed(3)} =`,
     real_time_factor.toFixed(3))

@@ -41,8 +41,8 @@ console.log('Done')
 const elapsed_seconds = (stop - start) / 1000;
 const duration = wave.samples.length / wave.sampleRate;
 const real_time_factor = elapsed_seconds / duration;
-console.log('Wave duration', duration.toFixed(3), 'secodns')
-console.log('Elapsed', elapsed_seconds.toFixed(3), 'secodns')
+console.log('Wave duration', duration.toFixed(3), 'seconds')
+console.log('Elapsed', elapsed_seconds.toFixed(3), 'seconds')
 console.log(
     `RTF = ${elapsed_seconds.toFixed(3)}/${duration.toFixed(3)} =`,
     real_time_factor.toFixed(3))

@@ -37,8 +37,8 @@ console.log('Done')
 const elapsed_seconds = (stop - start) / 1000;
 const duration = wave.samples.length / wave.sampleRate;
 const real_time_factor = elapsed_seconds / duration;
-console.log('Wave duration', duration.toFixed(3), 'secodns')
-console.log('Elapsed', elapsed_seconds.toFixed(3), 'secodns')
+console.log('Wave duration', duration.toFixed(3), 'seconds')
+console.log('Elapsed', elapsed_seconds.toFixed(3), 'seconds')
 console.log(
     `RTF = ${elapsed_seconds.toFixed(3)}/${duration.toFixed(3)} =`,
     real_time_factor.toFixed(3))

@@ -44,8 +44,8 @@ console.log('Done')
 const elapsed_seconds = (stop - start) / 1000;
 const duration = wave.samples.length / wave.sampleRate;
 const real_time_factor = elapsed_seconds / duration;
-console.log('Wave duration', duration.toFixed(3), 'secodns')
-console.log('Elapsed', elapsed_seconds.toFixed(3), 'secodns')
+console.log('Wave duration', duration.toFixed(3), 'seconds')
+console.log('Elapsed', elapsed_seconds.toFixed(3), 'seconds')
 console.log(
     `RTF = ${elapsed_seconds.toFixed(3)}/${duration.toFixed(3)} =`,
     real_time_factor.toFixed(3))

@@ -47,8 +47,8 @@ console.log('Done')
 const elapsed_seconds = (stop - start) / 1000;
 const duration = wave.samples.length / wave.sampleRate;
 const real_time_factor = elapsed_seconds / duration;
-console.log('Wave duration', duration.toFixed(3), 'secodns')
-console.log('Elapsed', elapsed_seconds.toFixed(3), 'secodns')
+console.log('Wave duration', duration.toFixed(3), 'seconds')
+console.log('Elapsed', elapsed_seconds.toFixed(3), 'seconds')
 console.log(
     `RTF = ${elapsed_seconds.toFixed(3)}/${duration.toFixed(3)} =`,
     real_time_factor.toFixed(3))

@@ -45,8 +45,8 @@ console.log('Done')
 const elapsed_seconds = (stop - start) / 1000;
 const duration = wave.samples.length / wave.sampleRate;
 const real_time_factor = elapsed_seconds / duration;
-console.log('Wave duration', duration.toFixed(3), 'secodns')
-console.log('Elapsed', elapsed_seconds.toFixed(3), 'secodns')
+console.log('Wave duration', duration.toFixed(3), 'seconds')
+console.log('Elapsed', elapsed_seconds.toFixed(3), 'seconds')
 console.log(
     `RTF = ${elapsed_seconds.toFixed(3)}/${duration.toFixed(3)} =`,
     real_time_factor.toFixed(3))

@@ -48,8 +48,8 @@ console.log('Done')
 const elapsed_seconds = (stop - start) / 1000;
 const duration = wave.samples.length / wave.sampleRate;
 const real_time_factor = elapsed_seconds / duration;
-console.log('Wave duration', duration.toFixed(3), 'secodns')
-console.log('Elapsed', elapsed_seconds.toFixed(3), 'secodns')
+console.log('Wave duration', duration.toFixed(3), 'seconds')
+console.log('Elapsed', elapsed_seconds.toFixed(3), 'seconds')
 console.log(
     `RTF = ${elapsed_seconds.toFixed(3)}/${duration.toFixed(3)} =`,
     real_time_factor.toFixed(3))

@@ -50,8 +50,8 @@ console.log('Done')
 const elapsed_seconds = (stop - start) / 1000;
 const duration = wave.samples.length / wave.sampleRate;
 const real_time_factor = elapsed_seconds / duration;
-console.log('Wave duration', duration.toFixed(3), 'secodns')
-console.log('Elapsed', elapsed_seconds.toFixed(3), 'secodns')
+console.log('Wave duration', duration.toFixed(3), 'seconds')
+console.log('Elapsed', elapsed_seconds.toFixed(3), 'seconds')
 console.log(
     `RTF = ${elapsed_seconds.toFixed(3)}/${duration.toFixed(3)} =`,
     real_time_factor.toFixed(3))

@@ -54,8 +54,8 @@ for (let filename of testWaves) {
   for (let e of events) {
     console.log(`${e.prob.toFixed(3)}\t\t\t${e.name}`);
   }
-  console.log('Wave duration', duration.toFixed(3), 'secodns')
-  console.log('Elapsed', elapsed_seconds.toFixed(3), 'secodns')
+  console.log('Wave duration', duration.toFixed(3), 'seconds')
+  console.log('Elapsed', elapsed_seconds.toFixed(3), 'seconds')
   console.log(
       `RTF = ${elapsed_seconds.toFixed(3)}/${duration.toFixed(3)} =`,
       real_time_factor.toFixed(3))

@@ -57,8 +57,8 @@ for (let filename of testWaves) {
   for (let e of events) {
     console.log(`${e.prob.toFixed(3)}\t\t\t${e.name}`);
   }
-  console.log('Wave duration', duration.toFixed(3), 'secodns')
-  console.log('Elapsed', elapsed_seconds.toFixed(3), 'secodns')
+  console.log('Wave duration', duration.toFixed(3), 'seconds')
+  console.log('Elapsed', elapsed_seconds.toFixed(3), 'seconds')
   console.log(
       `RTF = ${elapsed_seconds.toFixed(3)}/${duration.toFixed(3)} =`,
       real_time_factor.toFixed(3))

@@ -55,8 +55,8 @@ console.log('Done')
 const elapsed_seconds = (stop - start) / 1000;
 const duration = wave.samples.length / wave.sampleRate;
 const real_time_factor = elapsed_seconds / duration;
-console.log('Wave duration', duration.toFixed(3), 'secodns')
-console.log('Elapsed', elapsed_seconds.toFixed(3), 'secodns')
+console.log('Wave duration', duration.toFixed(3), 'seconds')
+console.log('Elapsed', elapsed_seconds.toFixed(3), 'seconds')
 console.log(
     `RTF = ${elapsed_seconds.toFixed(3)}/${duration.toFixed(3)} =`,
     real_time_factor.toFixed(3))

@@ -34,8 +34,8 @@ let stop = Date.now();
 const elapsed_seconds = (stop - start) / 1000;
 const duration = audio.samples.length / audio.sampleRate;
 const real_time_factor = elapsed_seconds / duration;
-console.log('Wave duration', duration.toFixed(3), 'secodns')
-console.log('Elapsed', elapsed_seconds.toFixed(3), 'secodns')
+console.log('Wave duration', duration.toFixed(3), 'seconds')
+console.log('Elapsed', elapsed_seconds.toFixed(3), 'seconds')
 console.log(
     `RTF = ${elapsed_seconds.toFixed(3)}/${duration.toFixed(3)} =`,
     real_time_factor.toFixed(3))

@@ -36,8 +36,8 @@ let stop = Date.now();
 const elapsed_seconds = (stop - start) / 1000;
 const duration = audio.samples.length / audio.sampleRate;
 const real_time_factor = elapsed_seconds / duration;
-console.log('Wave duration', duration.toFixed(3), 'secodns')
-console.log('Elapsed', elapsed_seconds.toFixed(3), 'secodns')
+console.log('Wave duration', duration.toFixed(3), 'seconds')
+console.log('Elapsed', elapsed_seconds.toFixed(3), 'seconds')
 console.log(
     `RTF = ${elapsed_seconds.toFixed(3)}/${duration.toFixed(3)} =`,
     real_time_factor.toFixed(3))

@@ -35,8 +35,8 @@ let stop = Date.now();
 const elapsed_seconds = (stop - start) / 1000;
 const duration = audio.samples.length / audio.sampleRate;
 const real_time_factor = elapsed_seconds / duration;
-console.log('Wave duration', duration.toFixed(3), 'secodns')
-console.log('Elapsed', elapsed_seconds.toFixed(3), 'secodns')
+console.log('Wave duration', duration.toFixed(3), 'seconds')
+console.log('Elapsed', elapsed_seconds.toFixed(3), 'seconds')
 console.log(
     `RTF = ${elapsed_seconds.toFixed(3)}/${duration.toFixed(3)} =`,
     real_time_factor.toFixed(3))

@@ -37,8 +37,8 @@ let stop = Date.now();
 const elapsed_seconds = (stop - start) / 1000;
 const duration = audio.samples.length / audio.sampleRate;
 const real_time_factor = elapsed_seconds / duration;
-console.log('Wave duration', duration.toFixed(3), 'secodns')
-console.log('Elapsed', elapsed_seconds.toFixed(3), 'secodns')
+console.log('Wave duration', duration.toFixed(3), 'seconds')
+console.log('Elapsed', elapsed_seconds.toFixed(3), 'seconds')
 console.log(
     `RTF = ${elapsed_seconds.toFixed(3)}/${duration.toFixed(3)} =`,
     real_time_factor.toFixed(3))

@@ -34,8 +34,8 @@ let stop = Date.now();
 const elapsed_seconds = (stop - start) / 1000;
 const duration = audio.samples.length / audio.sampleRate;
 const real_time_factor = elapsed_seconds / duration;
-console.log('Wave duration', duration.toFixed(3), 'secodns')
-console.log('Elapsed', elapsed_seconds.toFixed(3), 'secodns')
+console.log('Wave duration', duration.toFixed(3), 'seconds')
+console.log('Elapsed', elapsed_seconds.toFixed(3), 'seconds')
 console.log(
     `RTF = ${elapsed_seconds.toFixed(3)}/${duration.toFixed(3)} =`,
     real_time_factor.toFixed(3))

@@ -32,8 +32,8 @@ let stop = Date.now();
 const elapsed_seconds = (stop - start) / 1000;
 const duration = audio.samples.length / audio.sampleRate;
 const real_time_factor = elapsed_seconds / duration;
-console.log('Wave duration', duration.toFixed(3), 'secodns')
-console.log('Elapsed', elapsed_seconds.toFixed(3), 'secodns')
+console.log('Wave duration', duration.toFixed(3), 'seconds')
+console.log('Elapsed', elapsed_seconds.toFixed(3), 'seconds')
 console.log(
     `RTF = ${elapsed_seconds.toFixed(3)}/${duration.toFixed(3)} =`,
     real_time_factor.toFixed(3))

@@ -34,8 +34,8 @@ let stop = Date.now();
 const elapsed_seconds = (stop - start) / 1000;
 const duration = audio.samples.length / audio.sampleRate;
 const real_time_factor = elapsed_seconds / duration;
-console.log('Wave duration', duration.toFixed(3), 'secodns')
-console.log('Elapsed', elapsed_seconds.toFixed(3), 'secodns')
+console.log('Wave duration', duration.toFixed(3), 'seconds')
+console.log('Elapsed', elapsed_seconds.toFixed(3), 'seconds')
 console.log(
     `RTF = ${elapsed_seconds.toFixed(3)}/${duration.toFixed(3)} =`,
     real_time_factor.toFixed(3))

@@ -34,8 +34,8 @@ let stop = Date.now();
 const elapsed_seconds = (stop - start) / 1000;
 const duration = audio.samples.length / audio.sampleRate;
 const real_time_factor = elapsed_seconds / duration;
-console.log('Wave duration', duration.toFixed(3), 'secodns')
-console.log('Elapsed', elapsed_seconds.toFixed(3), 'secodns')
+console.log('Wave duration', duration.toFixed(3), 'seconds')
+console.log('Elapsed', elapsed_seconds.toFixed(3), 'seconds')
 console.log(
     `RTF = ${elapsed_seconds.toFixed(3)}/${duration.toFixed(3)} =`,
     real_time_factor.toFixed(3))