feat(engine): 重构字幕引擎并实现 WebSocket 通信

- 重构了 Gummy 和 Vosk 字幕引擎的代码，提高了可扩展性和可读性
- 合并 Gummy 和 Vosk 引擎为单个可执行文件
- 实现了字幕引擎和主程序之间的 WebSocket 通信，避免了孤儿进程问题

engine/main.py

@@ -1,10 +1,59 @@
 import argparse
+from utils import stdout_cmd
+from utils import thread_data, start_server
+from utils import merge_chunk_channels, resample_chunk_mono
+from audio2text import InvalidParameter, GummyRecognizer
+from audio2text import VoskRecognizer
+from sysaudio import AudioStream
 
-def gummy_engine(s, t, a, c, k):
-    pass
 
-def vosk_engine(a, c, m):
-    pass
+def main_gummy(s: str, t: str, a: int, c: int, k: str):
+    stream = AudioStream(a, c)
+    if t == 'none':
+        engine = GummyRecognizer(stream.RATE, s, None, k)
+    else:
+        engine = GummyRecognizer(stream.RATE, s, t, k)
+
+    stream.open_stream()
+    engine.start()
+
+    while thread_data.status == "running":
+        try:
+            chunk = stream.read_chunk()
+            if chunk is None: continue
+            chunk_mono = merge_chunk_channels(chunk, stream.CHANNELS)
+            try:
+                engine.send_audio_frame(chunk_mono)
+            except InvalidParameter:
+                stdout_cmd('info', 'Gummy engine stopped, restart engine')
+                engine.start()
+                engine.send_audio_frame(chunk_mono)
+        except KeyboardInterrupt:
+            break
+
+    stream.close_stream()
+    engine.stop()
+
+
+def main_vosk(a: int, c: int, m: str):
+    stream = AudioStream(a, c)
+    engine = VoskRecognizer(m)
+
+    stream.open_stream()
+    engine.start()
+
+    while thread_data.status == "running":
+        try:
+            chunk = stream.read_chunk()
+            if chunk is None: continue
+            chunk_mono = resample_chunk_mono(chunk, stream.CHANNELS, stream.RATE, 16000)
+            engine.send_audio_frame(chunk_mono)
+        except KeyboardInterrupt:
+            break
+
+    stream.close_stream()
+    engine.stop()
+
 
 if __name__ == "__main__":
     parser = argparse.ArgumentParser(description='Convert system audio stream to text')
@@ -12,15 +61,23 @@ if __name__ == "__main__":
     parser.add_argument('-e', '--caption_engine', default='gummy', help='Caption engine: gummy or vosk')
     parser.add_argument('-a', '--audio_type', default=0, help='Audio stream source: 0 for output, 1 for input')
     parser.add_argument('-c', '--chunk_rate', default=20, help='Number of audio stream chunks collected per second')
+    parser.add_argument('-p', '--port', default=7070, help='The port to run the server on, 0 for no server')
     # gummy
     parser.add_argument('-s', '--source_language', default='en', help='Source language code')
     parser.add_argument('-t', '--target_language', default='zh', help='Target language code')
     parser.add_argument('-k', '--api_key', default='', help='API KEY for Gummy model')
     # vosk
     parser.add_argument('-m', '--model_path', default='', help='The path to the vosk model.')
+    # for test
     args = parser.parse_args()
+    
+    if int(args.port) == 0:
+        thread_data.status = "running"
+    else:
+        start_server(int(args.port))
+
     if args.caption_engine == 'gummy':
-        gummy_engine(
+        main_gummy(
             args.source_language,
             args.target_language,
             int(args.audio_type),
@@ -28,7 +85,7 @@ if __name__ == "__main__":
             args.api_key
         )
     elif args.caption_engine == 'vosk':
-        vosk_engine(
+        main_vosk(
             int(args.audio_type),
             int(args.chunk_rate),
             args.model_path