Update app.py

app.py

@@ -5,30 +5,25 @@ subprocess.run(["pip", "install", "gradio", "--upgrade"])
 subprocess.run(["pip", "install", "datasets"])
 subprocess.run(["pip", "install", "transformers"])
 subprocess.run(["pip", "install", "torch", "torchvision", "torchaudio", "-f", "https://download.pytorch.org/whl/torch_stable.html"])
-import gradio as gr
-import numpy as np
-import torch
 from transformers import WhisperProcessor, WhisperForConditionalGeneration
+from datasets import load_dataset
 
-# Load model and processor
-processor = WhisperProcessor.from_pretrained("openai/whisper-large")
-model = WhisperForConditionalGeneration.from_pretrained("openai/whisper-large")
+# 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")
 
-# Custom preprocessing function
-def preprocess_audio(audio_data, sampling_rate=16_000):
-    sample_rate, raw_audio = audio_data
-    raw_speech = np.asarray(raw_audio, dtype=np.float32)
-    return {"input_values": raw_speech, "sampling_rate": sample_rate}
+# load dummy dataset and read audio files
+ds = load_dataset("hf-internal-testing/librispeech_asr_dummy", "clean", split="validation")
+sample = ds[0]["audio"]
+input_features = processor(sample["array"], sampling_rate=sample["sampling_rate"], return_tensors="pt").input_features 
 
-# Function to perform ASR on audio data
-def transcribe_audio(audio_data):
-    input_features = preprocess_audio(audio_data)
-    input_values = torch.tensor(input_features["input_values"]).unsqueeze(0)  # Add batch dimension
-    input_values = input_values.view(1, -1)  # Flatten the tensor to 2D
-    predicted_ids = model.generate(input_values)
-    transcription = processor.batch_decode(predicted_ids, skip_special_tokens=True)
-    return transcription[0]
+# generate token ids
+predicted_ids = model.generate(input_features)
+# decode token ids to text
+transcription = processor.batch_decode(predicted_ids, skip_special_tokens=False)
+
+transcription = processor.batch_decode(predicted_ids, skip_special_tokens=True)
 
 # Create Gradio interface
 audio_input = gr.Audio()