Onnx feature extractors
Collection
Various models in onnx format
•
2 items
•
Updated
MFCC feature extractor exported to onnx
sess = ort.InferenceSession(model_path)
input_name = sess.get_inputs()[0].name
output_name = sess.get_outputs()[0].name
mfcc_np = sess.run([output_name], {input_name: wav_np})[0]
mfcc_tensor = torch.from_numpy(feats_np).to(torch.float32).squeeze(0)