A dictionary learning and source recovery based approach to classify diverse audio sources
This work addresses audio source classification, which is important for applications like audio processing and recognition, but it appears incremental as it builds on existing dictionary learning techniques.
The paper tackled the problem of classifying audio signals among multiple sources using a dictionary learning approach, achieving a frame-wise classification accuracy of 98.2% for twelve sources and 100% accuracy for ten sources with a moving SDR method.
A dictionary learning based audio source classification algorithm is proposed to classify a sample audio signal as one amongst a finite set of different audio sources. Cosine similarity measure is used to select the atoms during dictionary learning. Based on three objective measures proposed, namely, signal to distortion ratio (SDR), the number of non-zero weights and the sum of weights, a frame-wise source classification accuracy of 98.2% is obtained for twelve different sources. Cent percent accuracy has been obtained using moving SDR accumulated over six successive frames for ten of the audio sources tested, while the two other sources require accumulation of 10 and 14 frames.