Add non-streaming ASR examples for Dart API (#1007)

dart-api-examples/.gitignore

@@ -0,0 +1 @@
+!run*.sh

dart-api-examples/non-streaming-asr/.gitignore

@@ -0,0 +1,3 @@
+# https://dart.dev/guides/libraries/private-files
+# Created by `dart pub`
+.dart_tool/

dart-api-examples/non-streaming-asr/CHANGELOG.md

@@ -0,0 +1,3 @@
+## 1.0.0
+
+- Initial version.

dart-api-examples/non-streaming-asr/README.md

@@ -0,0 +1,14 @@
+# Introduction
+
+This folder contains examples for non-streaming ASR with Dart API.
+
+| File | Description|
+|------|------------|
+|[./bin/nemo-ctc.dart](./bin/nemo-ctc.dart)| Use a NeMo Ctc model for speech recognition. See [./run-nemo-ctc.sh](./run-nemo-ctc.sh)|
+|[./bin/nemo-transducer.dart](./bin/nemo-transducer.dart)| Use a NeMo transducer model for speech recognition. See [./run-nemo-transducer.sh](./run-nemo-transducer.sh)|
+|[./bin/paraformer.dart](./bin/paraformer.dart)|Use a paraformer model for speech recognition. See [./run-paraformer.sh](./run-paraformer.sh)|
+|[./bin/telespeech-ctc.dart](./bin/telespeech-ctc.dart)| Use models from [Tele-AI/TeleSpeech-ASR](https://github.com/Tele-AI/TeleSpeech-ASR) for speech recognition. See [./run-telespeech-ctc.sh](./run-telespeech-ctc.sh)|
+|[./bin/whisper.dart](./bin/whisper.dart)| Use whisper for speech recognition. See [./run-whisper.sh](./run-whisper.sh)|
+|[./bin/zipformer-transducer.dart](./bin/zipformer-transducer.dart)| Use a zipformer transducer for speech recognition. See [./run-zipformer-transducer.sh](./run-zipformer-transducer.sh)|
+|[./bin/vad-with-paraformer.dart](./bin/vad-with-paraformer.dart)| Use a [silero-vad](https://github.com/snakers4/silero-vad) with paraformer for speech recognition. See [./run-vad-with-paraformer.sh](./run-vad-with-paraformer.sh)|
+

dart-api-examples/non-streaming-asr/analysis_options.yaml

@@ -0,0 +1,30 @@
+# This file configures the static analysis results for your project (errors,
+# warnings, and lints).
+#
+# This enables the 'recommended' set of lints from `package:lints`.
+# This set helps identify many issues that may lead to problems when running
+# or consuming Dart code, and enforces writing Dart using a single, idiomatic
+# style and format.
+#
+# If you want a smaller set of lints you can change this to specify
+# 'package:lints/core.yaml'. These are just the most critical lints
+# (the recommended set includes the core lints).
+# The core lints are also what is used by pub.dev for scoring packages.
+
+include: package:lints/recommended.yaml
+
+# Uncomment the following section to specify additional rules.
+
+# linter:
+#   rules:
+#     - camel_case_types
+
+# analyzer:
+#   exclude:
+#     - path/to/excluded/files/**
+
+# For more information about the core and recommended set of lints, see
+# https://dart.dev/go/core-lints
+
+# For additional information about configuring this file, see
+# https://dart.dev/guides/language/analysis-options

dart-api-examples/non-streaming-asr/bin/init.dart

@@ -0,0 +1 @@
+../../vad/bin/init.dart

dart-api-examples/non-streaming-asr/bin/nemo-ctc.dart

@@ -0,0 +1,52 @@
+import 'dart:io';
+import 'dart:typed_data';
+
+import 'package:args/args.dart';
+import 'package:sherpa_onnx/sherpa_onnx.dart' as sherpa_onnx;
+
+import './init.dart';
+
+void main(List<String> arguments) async {
+  await initSherpaOnnx();
+
+  final parser = ArgParser()
+    ..addOption('model', help: 'Path to the NeMo CTC model')
+    ..addOption('tokens', help: 'Path to tokens.txt')
+    ..addOption('input-wav', help: 'Path to input.wav to transcribe');
+
+  final res = parser.parse(arguments);
+  if (res['model'] == null ||
+      res['tokens'] == null ||
+      res['input-wav'] == null) {
+    print(parser.usage);
+    exit(1);
+  }
+
+  final model = res['model'] as String;
+  final tokens = res['tokens'] as String;
+  final inputWav = res['input-wav'] as String;
+
+  final nemo = sherpa_onnx.OfflineNemoEncDecCtcModelConfig(model: model);
+
+  final modelConfig = sherpa_onnx.OfflineModelConfig(
+    nemoCtc: nemo,
+    tokens: tokens,
+    debug: true,
+    numThreads: 1,
+  );
+  final config = sherpa_onnx.OfflineRecognizerConfig(model: modelConfig);
+  final recognizer = sherpa_onnx.OfflineRecognizer(config);
+
+  final waveData = sherpa_onnx.readWave(inputWav);
+  final stream = recognizer.createStream();
+
+  stream.acceptWaveform(
+      samples: waveData.samples, sampleRate: waveData.sampleRate);
+  recognizer.decode(stream);
+
+  final result = recognizer.getResult(stream);
+  print(result.text);
+
+  stream.free();
+  recognizer.free();
+}

dart-api-examples/non-streaming-asr/bin/nemo-transducer.dart

@@ -0,0 +1,62 @@
+import 'dart:io';
+import 'dart:typed_data';
+
+import 'package:args/args.dart';
+import 'package:sherpa_onnx/sherpa_onnx.dart' as sherpa_onnx;
+
+import './init.dart';
+
+void main(List<String> arguments) async {
+  await initSherpaOnnx();
+
+  final parser = ArgParser()
+    ..addOption('encoder', help: 'Path to the encoder model')
+    ..addOption('decoder', help: 'Path to decoder model')
+    ..addOption('joiner', help: 'Path to joiner model')
+    ..addOption('tokens', help: 'Path to tokens.txt')
+    ..addOption('input-wav', help: 'Path to input.wav to transcribe');
+
+  final res = parser.parse(arguments);
+  if (res['encoder'] == null ||
+      res['decoder'] == null ||
+      res['joiner'] == null ||
+      res['tokens'] == null ||
+      res['input-wav'] == null) {
+    print(parser.usage);
+    exit(1);
+  }
+
+  final encoder = res['encoder'] as String;
+  final decoder = res['decoder'] as String;
+  final joiner = res['joiner'] as String;
+  final tokens = res['tokens'] as String;
+  final inputWav = res['input-wav'] as String;
+
+  final transducer = sherpa_onnx.OfflineTransducerModelConfig(
+    encoder: encoder,
+    decoder: decoder,
+    joiner: joiner,
+  );
+
+  final modelConfig = sherpa_onnx.OfflineModelConfig(
+    transducer: transducer,
+    tokens: tokens,
+    debug: true,
+    numThreads: 1,
+  );
+  final config = sherpa_onnx.OfflineRecognizerConfig(model: modelConfig);
+  final recognizer = sherpa_onnx.OfflineRecognizer(config);
+
+  final waveData = sherpa_onnx.readWave(inputWav);
+  final stream = recognizer.createStream();
+
+  stream.acceptWaveform(
+      samples: waveData.samples, sampleRate: waveData.sampleRate);
+  recognizer.decode(stream);
+
+  final result = recognizer.getResult(stream);
+  print(result.text);
+
+  stream.free();
+  recognizer.free();
+}

dart-api-examples/non-streaming-asr/bin/paraformer.dart

@@ -0,0 +1,55 @@
+import 'dart:io';
+import 'dart:typed_data';
+
+import 'package:args/args.dart';
+import 'package:sherpa_onnx/sherpa_onnx.dart' as sherpa_onnx;
+
+import './init.dart';
+
+void main(List<String> arguments) async {
+  await initSherpaOnnx();
+
+  final parser = ArgParser()
+    ..addOption('model', help: 'Path to the paraformer model')
+    ..addOption('tokens', help: 'Path to tokens.txt')
+    ..addOption('input-wav', help: 'Path to input.wav to transcribe');
+
+  final res = parser.parse(arguments);
+  if (res['model'] == null ||
+      res['tokens'] == null ||
+      res['input-wav'] == null) {
+    print(parser.usage);
+    exit(1);
+  }
+
+  final model = res['model'] as String;
+  final tokens = res['tokens'] as String;
+  final inputWav = res['input-wav'] as String;
+
+  final paraformer = sherpa_onnx.OfflineParaformerModelConfig(
+    model: model,
+  );
+
+  final modelConfig = sherpa_onnx.OfflineModelConfig(
+    paraformer: paraformer,
+    tokens: tokens,
+    debug: true,
+    numThreads: 1,
+    modelType: 'paraformer',
+  );
+  final config = sherpa_onnx.OfflineRecognizerConfig(model: modelConfig);
+  final recognizer = sherpa_onnx.OfflineRecognizer(config);
+
+  final waveData = sherpa_onnx.readWave(inputWav);
+  final stream = recognizer.createStream();
+
+  stream.acceptWaveform(
+      samples: waveData.samples, sampleRate: waveData.sampleRate);
+  recognizer.decode(stream);
+
+  final result = recognizer.getResult(stream);
+  print(result.text);
+
+  stream.free();
+  recognizer.free();
+}

dart-api-examples/non-streaming-asr/bin/telespeech-ctc.dart

@@ -0,0 +1,51 @@
+import 'dart:io';
+import 'dart:typed_data';
+
+import 'package:args/args.dart';
+import 'package:sherpa_onnx/sherpa_onnx.dart' as sherpa_onnx;
+
+import './init.dart';
+
+void main(List<String> arguments) async {
+  await initSherpaOnnx();
+
+  final parser = ArgParser()
+    ..addOption('model', help: 'Path to the telespeech CTC model')
+    ..addOption('tokens', help: 'Path to tokens.txt')
+    ..addOption('input-wav', help: 'Path to input.wav to transcribe');
+
+  final res = parser.parse(arguments);
+  if (res['model'] == null ||
+      res['tokens'] == null ||
+      res['input-wav'] == null) {
+    print(parser.usage);
+    exit(1);
+  }
+
+  final model = res['model'] as String;
+  final tokens = res['tokens'] as String;
+  final inputWav = res['input-wav'] as String;
+
+  final modelConfig = sherpa_onnx.OfflineModelConfig(
+    telespeechCtc: model,
+    tokens: tokens,
+    debug: true,
+    numThreads: 1,
+    modelType: 'telespeech_ctc',
+  );
+  final config = sherpa_onnx.OfflineRecognizerConfig(model: modelConfig);
+  final recognizer = sherpa_onnx.OfflineRecognizer(config);
+
+  final waveData = sherpa_onnx.readWave(inputWav);
+  final stream = recognizer.createStream();
+
+  stream.acceptWaveform(
+      samples: waveData.samples, sampleRate: waveData.sampleRate);
+  recognizer.decode(stream);
+
+  final result = recognizer.getResult(stream);
+  print(result.text);
+
+  stream.free();
+  recognizer.free();
+}

dart-api-examples/non-streaming-asr/bin/vad-with-paraformer.dart

@@ -0,0 +1,97 @@
+import 'dart:io';
+import 'dart:typed_data';
+
+import 'package:args/args.dart';
+import 'package:sherpa_onnx/sherpa_onnx.dart' as sherpa_onnx;
+
+import './init.dart';
+
+void main(List<String> arguments) async {
+  await initSherpaOnnx();
+
+  final parser = ArgParser()
+    ..addOption('silero-vad', help: 'Path to silero_vad.onnx')
+    ..addOption('model', help: 'Path to the paraformer model')
+    ..addOption('tokens', help: 'Path to tokens.txt')
+    ..addOption('input-wav', help: 'Path to input.wav to transcribe');
+
+  final res = parser.parse(arguments);
+  if (res['silero-vad'] == null ||
+      res['model'] == null ||
+      res['tokens'] == null ||
+      res['input-wav'] == null) {
+    print(parser.usage);
+    exit(1);
+  }
+
+  final sileroVad = res['silero-vad'] as String;
+  final model = res['model'] as String;
+  final tokens = res['tokens'] as String;
+  final inputWav = res['input-wav'] as String;
+
+  final paraformer = sherpa_onnx.OfflineParaformerModelConfig(
+    model: model,
+  );
+
+  final modelConfig = sherpa_onnx.OfflineModelConfig(
+    paraformer: paraformer,
+    tokens: tokens,
+    debug: true,
+    numThreads: 1,
+    modelType: 'paraformer',
+  );
+  final config = sherpa_onnx.OfflineRecognizerConfig(model: modelConfig);
+  final recognizer = sherpa_onnx.OfflineRecognizer(config);
+
+  final sileroVadConfig = sherpa_onnx.SileroVadModelConfig(
+    model: sileroVad,
+    minSilenceDuration: 0.25,
+    minSpeechDuration: 0.5,
+  );
+
+  final vadConfig = sherpa_onnx.VadModelConfig(
+    sileroVad: sileroVadConfig,
+    numThreads: 1,
+    debug: true,
+  );
+
+  final vad = sherpa_onnx.VoiceActivityDetector(
+      config: vadConfig, bufferSizeInSeconds: 10);
+
+  final waveData = sherpa_onnx.readWave(inputWav);
+
+  int numSamples = waveData.samples.length;
+  int numIter = numSamples ~/ vadConfig.sileroVad.windowSize;
+
+  for (int i = 0; i != numIter; ++i) {
+    int start = i * vadConfig.sileroVad.windowSize;
+    vad.acceptWaveform(Float32List.sublistView(
+        waveData.samples, start, start + vadConfig.sileroVad.windowSize));
+
+    if (vad.isDetected()) {
+      while (!vad.isEmpty()) {
+        final stream = recognizer.createStream();
+        final segment = vad.front();
+        stream.acceptWaveform(
+            samples: segment.samples, sampleRate: waveData.sampleRate);
+        recognizer.decode(stream);
+
+        final result = recognizer.getResult(stream);
+
+        final startTime = segment.start * 1.0 / waveData.sampleRate;
+        final duration = segment.samples.length * 1.0 / waveData.sampleRate;
+        final stopTime = startTime + duration;
+        if (result.text != '') {
+          print(
+              '${startTime.toStringAsPrecision(4)} -- ${stopTime.toStringAsPrecision(4)}: ${result.text}');
+        }
+
+        stream.free();
+        vad.pop();
+      }
+    }
+  }
+
+  vad.free();
+  recognizer.free();
+}

dart-api-examples/non-streaming-asr/bin/whisper.dart

@@ -0,0 +1,59 @@
+import 'dart:io';
+import 'dart:typed_data';
+
+import 'package:args/args.dart';
+import 'package:sherpa_onnx/sherpa_onnx.dart' as sherpa_onnx;
+
+import './init.dart';
+
+void main(List<String> arguments) async {
+  await initSherpaOnnx();
+
+  final parser = ArgParser()
+    ..addOption('encoder', help: 'Path to the whisper encoder model')
+    ..addOption('decoder', help: 'Path to whisper decoder model')
+    ..addOption('tokens', help: 'Path to tokens.txt')
+    ..addOption('input-wav', help: 'Path to input.wav to transcribe');
+
+  final res = parser.parse(arguments);
+  if (res['encoder'] == null ||
+      res['decoder'] == null ||
+      res['tokens'] == null ||
+      res['input-wav'] == null) {
+    print(parser.usage);
+    exit(1);
+  }
+
+  final encoder = res['encoder'] as String;
+  final decoder = res['decoder'] as String;
+  final tokens = res['tokens'] as String;
+  final inputWav = res['input-wav'] as String;
+
+  final whisper = sherpa_onnx.OfflineWhisperModelConfig(
+    encoder: encoder,
+    decoder: decoder,
+  );
+
+  final modelConfig = sherpa_onnx.OfflineModelConfig(
+    whisper: whisper,
+    tokens: tokens,
+    modelType: 'whisper',
+    debug: false,
+    numThreads: 1,
+  );
+  final config = sherpa_onnx.OfflineRecognizerConfig(model: modelConfig);
+  final recognizer = sherpa_onnx.OfflineRecognizer(config);
+
+  final waveData = sherpa_onnx.readWave(inputWav);
+  final stream = recognizer.createStream();
+
+  stream.acceptWaveform(
+      samples: waveData.samples, sampleRate: waveData.sampleRate);
+  recognizer.decode(stream);
+
+  final result = recognizer.getResult(stream);
+  print(result.text);
+
+  stream.free();
+  recognizer.free();
+}

dart-api-examples/non-streaming-asr/bin/zipformer-transducer.dart

@@ -0,0 +1,62 @@
+import 'dart:io';
+import 'dart:typed_data';
+
+import 'package:args/args.dart';
+import 'package:sherpa_onnx/sherpa_onnx.dart' as sherpa_onnx;
+
+import './init.dart';
+
+void main(List<String> arguments) async {
+  await initSherpaOnnx();
+
+  final parser = ArgParser()
+    ..addOption('encoder', help: 'Path to the encoder model')
+    ..addOption('decoder', help: 'Path to decoder model')
+    ..addOption('joiner', help: 'Path to joiner model')
+    ..addOption('tokens', help: 'Path to tokens.txt')
+    ..addOption('input-wav', help: 'Path to input.wav to transcribe');
+
+  final res = parser.parse(arguments);
+  if (res['encoder'] == null ||
+      res['decoder'] == null ||
+      res['joiner'] == null ||
+      res['tokens'] == null ||
+      res['input-wav'] == null) {
+    print(parser.usage);
+    exit(1);
+  }
+
+  final encoder = res['encoder'] as String;
+  final decoder = res['decoder'] as String;
+  final joiner = res['joiner'] as String;
+  final tokens = res['tokens'] as String;
+  final inputWav = res['input-wav'] as String;
+
+  final transducer = sherpa_onnx.OfflineTransducerModelConfig(
+    encoder: encoder,
+    decoder: decoder,
+    joiner: joiner,
+  );
+
+  final modelConfig = sherpa_onnx.OfflineModelConfig(
+    transducer: transducer,
+    tokens: tokens,
+    debug: true,
+    numThreads: 1,
+  );
+  final config = sherpa_onnx.OfflineRecognizerConfig(model: modelConfig);
+  final recognizer = sherpa_onnx.OfflineRecognizer(config);
+
+  final waveData = sherpa_onnx.readWave(inputWav);
+  final stream = recognizer.createStream();
+
+  stream.acceptWaveform(
+      samples: waveData.samples, sampleRate: waveData.sampleRate);
+  recognizer.decode(stream);
+
+  final result = recognizer.getResult(stream);
+  print(result.text);
+
+  stream.free();
+  recognizer.free();
+}

dart-api-examples/non-streaming-asr/pubspec.lock

@@ -0,0 +1,96 @@
+# Generated by pub
+# See https://dart.dev/tools/pub/glossary#lockfile
+packages:
+  args:
+    dependency: "direct main"
+    description:
+      name: args
+      sha256: "7cf60b9f0cc88203c5a190b4cd62a99feea42759a7fa695010eb5de1c0b2252a"
+      url: "https://pub.dev"
+    source: hosted
+    version: "2.5.0"
+  characters:
+    dependency: transitive
+    description:
+      name: characters
+      sha256: "04a925763edad70e8443c99234dc3328f442e811f1d8fd1a72f1c8ad0f69a605"
+      url: "https://pub.dev"
+    source: hosted
+    version: "1.3.0"
+  collection:
+    dependency: transitive
+    description:
+      name: collection
+      sha256: ee67cb0715911d28db6bf4af1026078bd6f0128b07a5f66fb2ed94ec6783c09a
+      url: "https://pub.dev"
+    source: hosted
+    version: "1.18.0"
+  ffi:
+    dependency: transitive
+    description:
+      name: ffi
+      sha256: "493f37e7df1804778ff3a53bd691d8692ddf69702cf4c1c1096a2e41b4779e21"
+      url: "https://pub.dev"
+    source: hosted
+    version: "2.1.2"
+  flutter:
+    dependency: transitive
+    description: flutter
+    source: sdk
+    version: "0.0.0"
+  lints:
+    dependency: "direct dev"
+    description:
+      name: lints
+      sha256: cbf8d4b858bb0134ef3ef87841abdf8d63bfc255c266b7bf6b39daa1085c4290
+      url: "https://pub.dev"
+    source: hosted
+    version: "3.0.0"
+  material_color_utilities:
+    dependency: transitive
+    description:
+      name: material_color_utilities
+      sha256: "0e0a020085b65b6083975e499759762399b4475f766c21668c4ecca34ea74e5a"
+      url: "https://pub.dev"
+    source: hosted
+    version: "0.8.0"
+  meta:
+    dependency: transitive
+    description:
+      name: meta
+      sha256: "7687075e408b093f36e6bbf6c91878cc0d4cd10f409506f7bc996f68220b9136"
+      url: "https://pub.dev"
+    source: hosted
+    version: "1.12.0"
+  path:
+    dependency: "direct main"
+    description:
+      name: path
+      sha256: "087ce49c3f0dc39180befefc60fdb4acd8f8620e5682fe2476afd0b3688bb4af"
+      url: "https://pub.dev"
+    source: hosted
+    version: "1.9.0"
+  sherpa_onnx:
+    dependency: "direct main"
+    description:
+      name: sherpa_onnx
+      sha256: e45894f81e7c854ca96d678bcab5303036e884a7c90e9a6c4ec04c7b1ee215a8
+      url: "https://pub.dev"
+    source: hosted
+    version: "1.9.29"
+  sky_engine:
+    dependency: transitive
+    description: flutter
+    source: sdk
+    version: "0.0.99"
+  vector_math:
+    dependency: transitive
+    description:
+      name: vector_math
+      sha256: "80b3257d1492ce4d091729e3a67a60407d227c27241d6927be0130c98e741803"
+      url: "https://pub.dev"
+    source: hosted
+    version: "2.1.4"
+sdks:
+  dart: ">=3.4.0 <4.0.0"
+  flutter: ">=3.3.0"

dart-api-examples/non-streaming-asr/pubspec.yaml

@@ -0,0 +1,18 @@
+name: non_streaming_asr
+description: >
+  This example demonstrates how to use the Dart API for Non-streaming speech recognition. Specifically, we use the following models as examples, whisper, zipformer, and paraformer.
+
+version: 1.0.0
+# repository: https://github.com/my_org/my_repo
+
+environment:
+  sdk: ^3.4.0
+
+# Add regular dependencies here.
+dependencies:
+  sherpa_onnx: ^1.9.29
+  path: ^1.9.0
+  args: ^2.5.0
+
+dev_dependencies:
+  lints: ^3.0.0

dart-api-examples/non-streaming-asr/run-nemo-ctc.sh

@@ -0,0 +1,17 @@
+#!/usr/bin/env bash
+
+set -ex
+
+dart pub get
+
+if [ ! -f ./sherpa-onnx-nemo-fast-conformer-ctc-be-de-en-es-fr-hr-it-pl-ru-uk-20k/tokens.txt ]; then
+  curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-nemo-fast-conformer-ctc-be-de-en-es-fr-hr-it-pl-ru-uk-20k.tar.bz2
+  tar xvf sherpa-onnx-nemo-fast-conformer-ctc-be-de-en-es-fr-hr-it-pl-ru-uk-20k.tar.bz2
+  rm sherpa-onnx-nemo-fast-conformer-ctc-be-de-en-es-fr-hr-it-pl-ru-uk-20k.tar.bz2
+fi
+
+dart run \
+  ./bin/nemo-ctc.dart \
+  --model ./sherpa-onnx-nemo-fast-conformer-ctc-be-de-en-es-fr-hr-it-pl-ru-uk-20k/model.onnx \
+  --tokens ./sherpa-onnx-nemo-fast-conformer-ctc-be-de-en-es-fr-hr-it-pl-ru-uk-20k/tokens.txt \
+  --input-wav ./sherpa-onnx-nemo-fast-conformer-ctc-be-de-en-es-fr-hr-it-pl-ru-uk-20k/test_wavs/de-german.wav

dart-api-examples/non-streaming-asr/run-nemo-transducer.sh

@@ -0,0 +1,20 @@
+#!/usr/bin/env bash
+
+set -ex
+
+dart pub get
+
+if [ ! -f ./sherpa-onnx-nemo-fast-conformer-transducer-be-de-en-es-fr-hr-it-pl-ru-uk-20k/tokens.txt ]; then
+  curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-nemo-fast-conformer-transducer-be-de-en-es-fr-hr-it-pl-ru-uk-20k.tar.bz2
+
+  tar xvf sherpa-onnx-nemo-fast-conformer-transducer-be-de-en-es-fr-hr-it-pl-ru-uk-20k.tar.bz2
+  rm sherpa-onnx-nemo-fast-conformer-transducer-be-de-en-es-fr-hr-it-pl-ru-uk-20k.tar.bz2
+fi
+
+dart run \
+  ./bin/nemo-transducer.dart \
+  --encoder ./sherpa-onnx-nemo-fast-conformer-transducer-be-de-en-es-fr-hr-it-pl-ru-uk-20k/encoder.onnx \
+  --decoder ./sherpa-onnx-nemo-fast-conformer-transducer-be-de-en-es-fr-hr-it-pl-ru-uk-20k/decoder.onnx \
+  --joiner ./sherpa-onnx-nemo-fast-conformer-transducer-be-de-en-es-fr-hr-it-pl-ru-uk-20k/joiner.onnx \
+  --tokens ./sherpa-onnx-nemo-fast-conformer-transducer-be-de-en-es-fr-hr-it-pl-ru-uk-20k/tokens.txt \
+  --input-wav ./sherpa-onnx-nemo-fast-conformer-transducer-be-de-en-es-fr-hr-it-pl-ru-uk-20k/test_wavs/de-german.wav

dart-api-examples/non-streaming-asr/run-paraformer.sh

@@ -0,0 +1,18 @@
+#!/usr/bin/env bash
+
+set -ex
+
+dart pub get
+
+if [ ! -f ./sherpa-onnx-paraformer-zh-2023-03-28/tokens.txt ]; then
+  curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-paraformer-zh-2023-03-28.tar.bz2
+
+  tar xvf sherpa-onnx-paraformer-zh-2023-03-28.tar.bz2
+  rm sherpa-onnx-paraformer-zh-2023-03-28.tar.bz2
+fi
+
+dart run \
+  ./bin/paraformer.dart \
+  --model ./sherpa-onnx-paraformer-zh-2023-03-28/model.int8.onnx \
+  --tokens ./sherpa-onnx-paraformer-zh-2023-03-28/tokens.txt \
+  --input-wav ./sherpa-onnx-paraformer-zh-2023-03-28/test_wavs/3-sichuan.wav

dart-api-examples/non-streaming-asr/run-telespeech-ctc.sh

@@ -0,0 +1,18 @@
+#!/usr/bin/env bash
+
+set -ex
+
+dart pub get
+
+if [ ! -f ./sherpa-onnx-telespeech-ctc-int8-zh-2024-06-04/tokens.txt ]; then
+  curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-telespeech-ctc-int8-zh-2024-06-04.tar.bz2
+
+  tar xvf sherpa-onnx-telespeech-ctc-int8-zh-2024-06-04.tar.bz2
+  rm sherpa-onnx-telespeech-ctc-int8-zh-2024-06-04.tar.bz2
+fi
+
+dart run \
+  ./bin/telespeech-ctc.dart \
+  --model ./sherpa-onnx-telespeech-ctc-int8-zh-2024-06-04/model.int8.onnx \
+  --tokens ./sherpa-onnx-telespeech-ctc-int8-zh-2024-06-04/tokens.txt \
+  --input-wav ./sherpa-onnx-telespeech-ctc-int8-zh-2024-06-04/test_wavs/3-sichuan.wav

dart-api-examples/non-streaming-asr/run-vad-with-paraformer.sh

@@ -0,0 +1,27 @@
+#!/usr/bin/env bash
+
+set -ex
+
+dart pub get
+
+if [[ ! -f ./silero_vad.onnx ]]; then
+  curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/silero_vad.onnx
+fi
+
+if [[ ! -f ./lei-jun-test.wav ]]; then
+  curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/lei-jun-test.wav
+fi
+
+if [ ! -f ./sherpa-onnx-paraformer-zh-2023-03-28/tokens.txt ]; then
+  curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-paraformer-zh-2023-03-28.tar.bz2
+
+  tar xvf sherpa-onnx-paraformer-zh-2023-03-28.tar.bz2
+  rm sherpa-onnx-paraformer-zh-2023-03-28.tar.bz2
+fi
+
+dart run \
+  ./bin/vad-with-paraformer.dart \
+  --silero-vad ./silero_vad.onnx \
+  --model ./sherpa-onnx-paraformer-zh-2023-03-28/model.int8.onnx \
+  --tokens ./sherpa-onnx-paraformer-zh-2023-03-28/tokens.txt \
+  --input-wav ./lei-jun-test.wav

dart-api-examples/non-streaming-asr/run-whisper.sh

@@ -0,0 +1,19 @@
+#!/usr/bin/env bash
+
+set -ex
+
+dart pub get
+
+if [ ! -f ./sherpa-onnx-whisper-tiny.en/tiny.en-tokens.txt ]; then
+  curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-whisper-tiny.en.tar.bz2
+
+  tar xvf sherpa-onnx-whisper-tiny.en.tar.bz2
+  rm sherpa-onnx-whisper-tiny.en.tar.bz2
+fi
+
+dart run \
+  ./bin/whisper.dart \
+  --encoder ./sherpa-onnx-whisper-tiny.en/tiny.en-encoder.int8.onnx \
+  --decoder ./sherpa-onnx-whisper-tiny.en/tiny.en-decoder.int8.onnx \
+  --tokens ./sherpa-onnx-whisper-tiny.en/tiny.en-tokens.txt \
+  --input-wav ./sherpa-onnx-whisper-tiny.en/test_wavs/0.wav

dart-api-examples/non-streaming-asr/run-zipformer-transducer.sh

@@ -0,0 +1,20 @@
+#!/usr/bin/env bash
+
+set -ex
+
+dart pub get
+
+if [ ! -f ./sherpa-onnx-zipformer-gigaspeech-2023-12-12/tokens.txt ]; then
+  curl -SL -O https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-zipformer-gigaspeech-2023-12-12.tar.bz2
+
+  tar xvf sherpa-onnx-zipformer-gigaspeech-2023-12-12.tar.bz2
+  rm sherpa-onnx-zipformer-gigaspeech-2023-12-12.tar.bz2
+fi
+
+dart run \
+  ./bin/zipformer-transducer.dart \
+  --encoder ./sherpa-onnx-zipformer-gigaspeech-2023-12-12/encoder-epoch-30-avg-1.int8.onnx \
+  --decoder ./sherpa-onnx-zipformer-gigaspeech-2023-12-12/decoder-epoch-30-avg-1.onnx \
+  --joiner ./sherpa-onnx-zipformer-gigaspeech-2023-12-12/joiner-epoch-30-avg-1.int8.onnx \
+  --tokens ./sherpa-onnx-zipformer-gigaspeech-2023-12-12/tokens.txt \
+  --input-wav ./sherpa-onnx-zipformer-gigaspeech-2023-12-12/test_wavs/1221-135766-0001.wav

dart-api-examples/vad/bin/init.dart

@@ -0,0 +1,29 @@
+import 'dart:io';
+import 'dart:isolate';
+import 'package:path/path.dart' as p;
+import 'package:sherpa_onnx/sherpa_onnx.dart' as sherpa_onnx;
+
+Future<void> initSherpaOnnx() async {
+  var uri = await Isolate.resolvePackageUri(
+      Uri.parse('package:sherpa_onnx/sherpa_onnx.dart'));
+
+  if (uri == null) {
+    print('File not found');
+    exit(1);
+  }
+
+  String platform = '';
+
+  if (Platform.isMacOS) {
+    platform = 'macos';
+  } else if (Platform.isLinux) {
+    platform = 'linux';
+  } else if (Platform.isWindows) {
+    platform = 'windows';
+  } else {
+    throw UnsupportedError('Unknown platform: ${Platform.operatingSystem}');
+  }
+
+  final libPath = p.join(p.dirname(p.fromUri(uri)), '..', platform);
+  sherpa_onnx.initBindings(libPath);
+}

dart-api-examples/vad/bin/vad.dart

@@ -1,33 +1,9 @@
 import 'dart:io';
-import 'dart:isolate';
 import 'dart:typed_data';
 
 import 'package:args/args.dart';
-import 'package:path/path.dart' as p;
 import 'package:sherpa_onnx/sherpa_onnx.dart' as sherpa_onnx;
-
-Future<void> initSherpaOnnx() async {
-  var uri = await Isolate.resolvePackageUri(
-      Uri.parse('package:sherpa_onnx/sherpa_onnx.dart'));
-
-  if (uri == null) {
-    print('File not found');
-    exit(1);
-  }
-  String platform = '';
-  if (Platform.isMacOS) {
-    platform = 'macos';
-  } else if (Platform.isLinux) {
-    platform = 'linux';
-  } else if (Platform.isWindows) {
-    platform = 'windows';
-  } else {
-    throw UnsupportedError('Unknown platform: ${Platform.operatingSystem}');
-  }
-
-  final libPath = p.join(p.dirname(p.fromUri(uri)), '..', platform);
-  sherpa_onnx.initBindings(libPath);
-}
+import './init.dart';
 
 void main(List<String> arguments) async {
   await initSherpaOnnx();
@@ -36,6 +12,7 @@ void main(List<String> arguments) async {
     ..addOption('silero-vad', help: 'Path to silero_vad.onnx')
     ..addOption('input-wav', help: 'Path to input.wav')
     ..addOption('output-wav', help: 'Path to output.wav');
+
   final res = parser.parse(arguments);
   if (res['silero-vad'] == null ||
       res['input-wav'] == null ||
@@ -53,6 +30,7 @@ void main(List<String> arguments) async {
     minSilenceDuration: 0.25,
     minSpeechDuration: 0.5,
   );
+
   final config = sherpa_onnx.VadModelConfig(
     sileroVad: sileroVadConfig,
     numThreads: 1,
@@ -86,8 +64,11 @@ void main(List<String> arguments) async {
     }
   }
 
+  vad.free();
+
   final s = Float32List.fromList(allSamples.expand((x) => x).toList());
   sherpa_onnx.writeWave(
       filename: outputWav, samples: s, sampleRate: waveData.sampleRate);
+
   print('Saved to ${outputWav}');
 }

dart-api-examples/vad/pubspec.lock

@@ -74,10 +74,10 @@ packages:
     dependency: "direct main"
     description:
       name: sherpa_onnx
-      sha256: "6cfadf7bc35001bb1284f9fac1e03e33787cafa918e0c45da96d1e91afa58751"
+      sha256: e45894f81e7c854ca96d678bcab5303036e884a7c90e9a6c4ec04c7b1ee215a8
       url: "https://pub.dev"
     source: hosted
-    version: "0.0.3"
+    version: "1.9.29"
   sky_engine:
     dependency: transitive
     description: flutter

dart-api-examples/vad/pubspec.yaml

@@ -9,7 +9,7 @@ environment:
   sdk: ^3.4.0
 
 dependencies:
-  sherpa_onnx: ^0.0.3
+  sherpa_onnx: ^1.9.29
   path: ^1.9.0
   args: ^2.5.0