zerogpu

app.py

@@ -7,6 +7,7 @@ import librosa
 import tempfile
 from twilio.rest import Client
 import os
+import spaces
 from transformers import Qwen2AudioForConditionalGeneration, AutoProcessor
 
 processor = AutoProcessor.from_pretrained("Qwen/Qwen2-Audio-7B-Instruct")
@@ -28,13 +29,9 @@ else:
     rtc_configuration = None
 
 
+@spaces.GPU
 def transcribe(audio: tuple[int, np.ndarray], transformers_convo: list[dict], gradio_convo: list[dict]):
-    segment = AudioSegment(
-        audio[1].tobytes(),
-        frame_rate=audio[0],
-        sample_width=audio[1].dtype.itemsize,
-        channels=1,
-    )
+    segment = AudioSegment(audio[1].tobytes(), frame_rate=audio[0], sample_width=audio[1].dtype.itemsize, channels=1)
 
     with tempfile.NamedTemporaryFile(suffix=".mp3") as temp_audio:
         segment.export(temp_audio.name, format="mp3")