Update app.py

app.py

@@ -8,24 +8,24 @@ subprocess.run(["pip", "install", "transformers"])
 subprocess.run(["pip", "install", "librosa", "soundfile"])
 subprocess.run(["pip", "install", "torch", "torchvision", "torchaudio", "-f", "https://download.pytorch.org/whl/torch_stable.html"])
 
+import gradio as gr
 from transformers import WhisperProcessor, WhisperForConditionalGeneration
-from datasets import load_dataset
+import numpy as np
 
-# Define the transcribe_audio function
-def transcribe_audio(audio):
-    input_features = processor(audio, return_tensors="pt").input_features 
-    predicted_ids = model.generate(input_features)
-    transcription = processor.batch_decode(predicted_ids, skip_special_tokens=True)
-    return transcription[0]
-
-# load model and processor
+# Load model and processor
 processor = WhisperProcessor.from_pretrained("openai/whisper-small")
 model = WhisperForConditionalGeneration.from_pretrained("openai/whisper-small")
 forced_decoder_ids = processor.get_decoder_prompt_ids(language="italian", task="transcribe")
 
-# load dummy dataset and read audio files
-ds = load_dataset("hf-internal-testing/librispeech_asr_dummy", "clean", split="validation")
-sample = ds[0]["audio"]
+def transcribe_audio(audio):
+    # Assuming sampling_rate is known
+    sampling_rate = 16000  # Change this to the actual sampling rate of your audio
+
+    # Ensure to pass the sampling_rate parameter
+    input_features = processor(audio, sampling_rate=sampling_rate, return_tensors="pt").input_features
+    predicted_ids = model.generate(input_features)
+    transcription = processor.batch_decode(predicted_ids, skip_special_tokens=True)
+    return transcription[0]
 
 # Create Gradio interface
 audio_input = gr.Audio()