ZeroGPU spaces

app.py

@@ -4,6 +4,7 @@ import torch
 import numpy as np
 import librosa
 from transformers import AutoProcessor, Wav2Vec2BertForCTC
+import spaces
 
 MODEL_NAME = "mikr/w2v-bert-2.0-czech-colab-cv16"
 
@@ -15,6 +16,7 @@ processor = AutoProcessor.from_pretrained(MODEL_NAME)
 model = Wav2Vec2BertForCTC.from_pretrained(MODEL_NAME).to(device)
 
 
+@spaces.GPU
 def transcribe(audio_path):
     a, s = librosa.load(audio_path, sr=16_000)
 

requirements.txt

@@ -3,3 +3,4 @@ torch
 soundfile
 librosa
 ffmpy
+spaces