Conference Papers

Permanent URI for this collectionhttps://idr.nitk.ac.in/handle/123456789/28506

Browse

Subject: search.filters.subject.Music Information Retrieval

Search Results

Now showing 1 - 2 of 2
  • No Thumbnail Available
    Item
    Audio segmentation using a priori information in the context of Karnatic Music
    (Institute of Electrical and Electronics Engineers Inc., 2015) Kalyan, V.A.; Sankaranarayanan, S.; Sumam David, S.
    Karnatic Music (KM) is distinct because of the prevalence of gamaka - embellishments to musical notes in the form of frequency traversals. Another important aspect of KM is that the performance style is mostly extempore. Hence, Music Information Retrieval (MIR) tasks in the context of KM are highly challenging. This paper deals with the task of Audio Segmentation and its application to MIR challenges of KM at various levels. This work presents a method that incorporates a priori knowledge about the music system and the audio track at hand for segmenting the audio into its constituent notes. The method uses amplitude and energy based features to train a neural network and an accuracy of 95.2% has been achieved on KM audio samples. The paper also elucidates the application of the method to important MIR tasks such as Music Transcription and Score-Alignment in the context of KM. © 2015 IEEE.
  • No Thumbnail Available
    Item
    Beat Onset Detection in an Audio Clip of a Percussion Instrument-Mridanga
    (Institute of Electrical and Electronics Engineers Inc., 2018) Vishnu Swaroop, G.; Koolagudi, S.G.; Vishnu Srinivasa Murthy, Y.V.
    The process of automating MIR tasks is essential due to the availability of enormous number of tracks. Of these, beat onset detection is a base for the task of Tala identification which is a part of Indian Classical Music (ICM). In this paper, an effort has been made to detect the onset of a specific percussion instrument called Mridanga as it is highly used instrument in Carnatic music. The dataset has been recorded at studio by playing the Mridanga for different Talas. Further, various signal to noise ratio (SNR) values have added in the range of 40 dB - 10 dB to generalize the system for real-time applications. The features such as centroid flux, and rate of change in energy have been computed for every sub-band. Various filtering approaches have been used to optimize the process of stroke onset detection. The results are found to be appreciable and an average accuracy of 95.01% is obtained with 40dB and 89.73% is with 10dB. Copy Right © INDIACom-2018.