Abstract not found

This paper contains a review of the issues surrounding sinusoidal parameter estimation which is a vital part of many audio manipulation algorithms. A number of algorithms which use the phase of the Fourier transform for estimation (e.g. [1]) are explored and shown to be identical. Their performance against a classical interpolation estimator [2] and comparison with the Cramer Rao Bound (CRB) is presented. Component detection is also considered and various methods of improving these algorithms are discussed.

This report centres around some of this issues involved in automatic transcription of polyphonic musical audio signals. That is, representing the information contained in the audio in such a way as to be recognisable and usable by a musician. First, a review of the various fields which have a bearing on the subject is put forward, including music, music psychology, auditory psychology and signal processing. Then a thorough appraisal of previous work on automated polyphonic transcription is presented. Next, original work on the use of time-frequency reassignment as a front end is imparted and finally, future ideas are expounded and a timetable for forthcoming research is given.

In extending previous work on detecting transient spectral peaks we here investigate into the distinction between sinusoidal and noise components by means of classification of spectral peaks. The classification is based on descriptors derived from properties related to time-frequency distributions. In contrast to existing methods, the descriptors are designed to properly deal with non-stationary sinusoids, which considerably increases the range of applications. The experimental investigation shows superior classification results compared to the standard correlation-based approach. 1

Time-frequency representations such as the spectrogram are commonly used to analyze signals having a time-varying distribution of spectral energy, but the spectrogram is constrained by an unfortunate tradeoff between resolution in time and frequency. A method of achieving high-resolution spectral representations has been independently introduced by several parties. The technique has been variously named reassignment and remapping, but while the implementations have differed in details, they are all based on the same theoretical and mathematical foundation. In this work, we present a brief history of work on the method we will call the method of time-frequency reassignment, and present a unified mathematical description of the technique and its derivation. We will focus on the development of time-frequency reassignment in the context of the spectrogram, and conclude with a discussion of some current applications of the reassigned spectrogram. Key words: Time-frequency representations, reassignment, spectral analysis PACS: 43.60.Hj, 43.60.Ac, 43.58.Kr 1