feat: integrate Sherpa-ONNX local offline speech recognition engine

2026-09-15 22:22:32 +08:00
parent db7ec7404f
commit be5075fa3d
12 changed files with 786 additions and 4 deletions
@@ -0,0 +1,49 @@
+import 'dart:io';
+import 'dart:typed_data';
+import 'package:flutter_test/flutter_test.dart';
+import 'package:kouyu_english/core/sherpa_stt_service.dart';
+import 'package:sherpa_onnx/sherpa_onnx.dart' as sherpa_onnx;
+
+void main() {
+  TestWidgetsFlutterBinding.ensureInitialized();
+
+  test('SherpaSttService transcribes WAV audio file accurately', () async {
+    const macosDir = '/Users/shen/.pub-cache/hosted/pub.dev/sherpa_onnx_macos-1.13.8/macos';
+    
+    // Test direct Sherpa ASR initialization and decoding with bundled test audio
+    sherpa_onnx.initBindings(macosDir);
+
+    final modelConfig = sherpa_onnx.OfflineModelConfig(
+      transducer: const sherpa_onnx.OfflineTransducerModelConfig(
+        encoder: 'assets/models/sherpa/encoder-epoch-99-avg-1.int8.onnx',
+        decoder: 'assets/models/sherpa/decoder-epoch-99-avg-1.int8.onnx',
+        joiner: 'assets/models/sherpa/joiner-epoch-99-avg-1.int8.onnx',
+      ),
+      tokens: 'assets/models/sherpa/tokens.txt',
+      numThreads: 2,
+      debug: false,
+    );
+
+    final recognizerConfig = sherpa_onnx.OfflineRecognizerConfig(
+      model: modelConfig,
+      feat: const sherpa_onnx.FeatureConfig(sampleRate: 16000, featureDim: 80),
+    );
+
+    final recognizer = sherpa_onnx.OfflineRecognizer(recognizerConfig);
+
+    const testWave = '/tmp/sherpa_test/sherpa-onnx-zipformer-small-en-2023-06-26/test_wavs/0.wav';
+    if (File(testWave).existsSync()) {
+      final wave = sherpa_onnx.readWave(testWave);
+      expect(wave.samples.isNotEmpty, isTrue);
+
+      final stream = recognizer.createStream();
+      stream.acceptWaveform(samples: wave.samples, sampleRate: wave.sampleRate);
+      recognizer.decode(stream);
+      final result = recognizer.getResult(stream);
+      print("Transcribed result: ${result.text}");
+      expect(result.text.contains("NIGHTFALL"), isTrue);
+      stream.free();
+    }
+    recognizer.free();
+  });
+}