edited

app.py

@@ -13,7 +13,7 @@ from pyannote.core import Segment
 from transformers.pipelines.audio_utils import ffmpeg_read
 
 
-MODEL_NAME = "large-v3"
+MODEL_NAME = "medium"
 BATCH_SIZE = 8
 FILE_LIMIT_MB = 1000
 YT_LENGTH_LIMIT_S = 3600  # limit to 1 hour YouTube files
@@ -36,8 +36,8 @@ def transcribe(inputs, task):
         transform = torchaudio.transforms.Resample(orig_freq=sample_rate, new_freq=16000)
         waveform = transform(waveform)
 
-    diarization = pipeline({"uri": "audio", "audio": audio_path})
-    speaker_segments = []
+    #diarization = pipeline({"uri": "audio", "audio": audio_path})
+    #speaker_segments = []
     return text
 
 def _return_yt_html_embed(yt_url):