Update app.py

app.py

@@ -1,17 +1,19 @@
-import os
+import spaces
 import torch
 import gradio as gr
 import whisperx
 from transformers.pipelines.audio_utils import ffmpeg_read
 import tempfile
 import gc
+import os
 
 # Constants
 DEVICE = "cuda" if torch.cuda.is_available() else "cpu"
-BATCH_SIZE = 4
-COMPUTE_TYPE = "float32"
+BATCH_SIZE = 4  # reduce if low on GPU mem
+COMPUTE_TYPE = "float32"  # change to "int8" if low on GPU mem
 FILE_LIMIT_MB = 1000
 
+@spaces.GPU
 def transcribe_audio(inputs, task):
     if inputs is None:
         raise gr.Error("No audio file submitted! Please upload or record an audio file before submitting your request.")
@@ -44,7 +46,7 @@ def transcribe_audio(inputs, task):
         torch.cuda.empty_cache()
         
         # 3. Diarize audio
-        diarize_model = whisperx.DiarizationPipeline(use_auth_token="YOUR_HF_TOKEN", device=DEVICE)
+        diarize_model = whisperx.DiarizationPipeline(use_auth_token=os.environ["HF_TOKEN"], device=DEVICE)
         diarize_segments = diarize_model(audio)
         
         # 4. Assign speaker labels
@@ -111,4 +113,4 @@ with demo:
         outputs=output_text
     )
 
-demo.queue().launch(share=True)
+demo.queue().launch(ssr_mode=False)