| # TEST MODEL | |
| from transformers import pipeline | |
| classifier = pipeline(task="zero-shot-audio-classification", model="mskov/whisper-small-esc50") | |
| # classifier = pipeline(model="mskov/roberta-base-toxicity") | |
| audio = "./candy-bar-chewing.wav" | |
| labels = ["Sound of a dog", "Sound of vaccum cleaner", "chewing", "sneezing"] | |
| result = [] | |
| for item in labels: | |
| result.append(classifier(audio, input_ids=labels)) | |
| predicted_label = result[0]["label"] | |
| print(f"Predicted label: {predicted_label}") |