Merge pull request #384 from klonikar/main

issue 371

run_client.py

@@ -0,0 +1,57 @@
+from pathlib import Path
+import sys
+from whisper_live.client import TranscriptionClient
+import argparse
+
+if __name__ == '__main__':
+  parser = argparse.ArgumentParser()
+  parser.add_argument('--port', '-p',
+                        type=int,
+                        default=9090,
+                        help="Websocket port to run the server on.")
+  parser.add_argument('--server', '-s',
+                        type=str,
+                        default='localhost',
+                        help='hostname or ip address of server')
+  parser.add_argument('--files', '-f',
+                        type=str,
+                        nargs='+',
+                        help='hostname or ip address of server')
+  parser.add_argument('--output_file', '-o',
+                        type=str,
+                        default='./output_recording.wav',
+                        help='hostname or ip address of server')
+  args = parser.parse_args()
+
+  # Validate audio files
+  valid_files = []
+  for file_path in args.files:
+      path = Path(file_path)
+      if path.exists() and path.is_file():
+          valid_files.append(str(path))
+      else:
+          print(f"Warning: File not found: {file_path}")
+
+  if not valid_files:
+      print("Error: No valid audio files found!")
+      sys.exit(1)
+
+  print(f"Found {len(valid_files)} audio file(s) to stream:")
+  for file_path in valid_files:
+      print(f"  - {file_path}")
+
+  for f in valid_files:
+    client = TranscriptionClient(
+      args.server,
+      args.port,
+      lang="en",
+      translate=False,
+      model="large-v3",                                   # also support hf_model => `Systran/faster-whisper-small`
+      use_vad=False,
+      save_output_recording=False,                         # Only used for microphone input, False by Default
+      output_recording_filename=args.output_file, # Only used for microphone input
+      max_clients=4,
+      max_connection_time=600,
+      mute_audio_playback=True,                          # Only used for file input, False by Default
+    )
+    client(f)

whisper_live/client.py

@@ -420,14 +420,18 @@ class TranscriptionTeeClient:
 
         # read audio and create pyaudio stream
         with wave.open(filename, "rb") as wavfile:
-            self.stream = self.p.open(
-                format=self.p.get_format_from_width(wavfile.getsampwidth()),
-                channels=wavfile.getnchannels(),
-                rate=wavfile.getframerate(),
-                input=True,
-                output=True,
-                frames_per_buffer=self.chunk,
-            )
+            if self.mute_audio_playback:
+                self.stream = None
+            else:
+                self.stream = self.p.open(
+                    format=self.p.get_format_from_width(wavfile.getsampwidth()),
+                    channels=wavfile.getnchannels(),
+                    rate=wavfile.getframerate(),
+                    input=True,
+                    output=True,
+                    frames_per_buffer=self.chunk,
+                )
+
             chunk_duration = self.chunk / float(wavfile.getframerate())
             try:
                 while any(client.recording for client in self.clients):
@@ -448,7 +452,8 @@ class TranscriptionTeeClient:
                     client.wait_before_disconnect()
                 self.multicast_packet(Client.END_OF_AUDIO.encode('utf-8'), True)
                 self.write_all_clients_srt()
-                self.stream.close()
+                if self.stream:
+                    self.stream.close()
                 self.close_all_clients()
 
             except KeyboardInterrupt:

whisper_live/server.py

@@ -216,7 +216,8 @@ class TranscriptionServer:
         try:
             if self.backend.is_faster_whisper():
                 from whisper_live.backend.faster_whisper_backend import ServeClientFasterWhisper
-                if faster_whisper_custom_model_path is not None and os.path.exists(faster_whisper_custom_model_path):
+                # model is of the form namespace/repo_name and not a filesystem path
+                if faster_whisper_custom_model_path is not None:
                     logging.info(f"Using custom model {faster_whisper_custom_model_path}")
                     options["model"] = faster_whisper_custom_model_path
                 client = ServeClientFasterWhisper(
@@ -380,8 +381,6 @@ class TranscriptionServer:
             port (int): The port number to bind the server.
         """
         self.cache_path = cache_path
-        if faster_whisper_custom_model_path is not None and not os.path.exists(faster_whisper_custom_model_path):
-            raise ValueError(f"Custom faster_whisper model '{faster_whisper_custom_model_path}' is not a valid path.")
         if whisper_tensorrt_path is not None and not os.path.exists(whisper_tensorrt_path):
             raise ValueError(f"TensorRT model '{whisper_tensorrt_path}' is not a valid path.")
         if single_model: