update feature for real time transcription locally

whisjax_realtime_trial.py

@@ -14,23 +14,28 @@ WHISPER_MODEL_SIZE = config['DEFAULT']["WHISPER_MODEL_SIZE"]
 
 FRAMES_PER_BUFFER = 8000
 FORMAT = pyaudio.paInt16
-CHANNELS = 1
+CHANNELS = 2
 RATE = 44100
-RECORD_SECONDS = 5
+RECORD_SECONDS = 15
 
 
 def main():
     p = pyaudio.PyAudio()
-
+    AUDIO_DEVICE_ID = -1
+    for i in range(p.get_device_count()):
+        if p.get_device_info_by_index(i)["name"] == "ref-agg-input":
+            AUDIO_DEVICE_ID = i
+    audio_devices = p.get_device_info_by_index(AUDIO_DEVICE_ID)
     stream = p.open(
         format=FORMAT,
         channels=CHANNELS,
         rate=RATE,
         input=True,
-        frames_per_buffer=FRAMES_PER_BUFFER
+        frames_per_buffer=FRAMES_PER_BUFFER,
+        input_device_index=audio_devices['index']
     )
 
-    pipeline = FlaxWhisperPipline("openai/whisper-" + WHISPER_MODEL_SIZE,
+    pipeline = FlaxWhisperPipline("openai/whisper-" + config["DEFAULT"]["WHISPER_REAL_TIME_MODEL_SIZE"],
                                   dtype=jnp.float16,
                                   batch_size=16)
 
@@ -48,8 +53,7 @@ def main():
 
     listener = keyboard.Listener(on_press=on_press)
     listener.start()
-    print("Listening...")
-
+    print("Attempting real-time transcription.. Listening...")
     while proceed:
         try:
             frames = []