A Robust Step Warping Method for Tone Sequences

Abstract

A robust step regulating method for tonal sequences is used for regulating inaccurate tonal values to tonal values more aligned with a context and has good effect on determining tonal values. The method includes the steps of extracting original tonal sequences from input audio digital signals, repeatedly intercepting sub-tonal sequences with a constant-length, mapping the sub-tonal sequences to a regulating matrix, expanding the regulating matrix to construct a corresponding graph structure, obtaining the path therein to generate a path associated matrix, and analyzing the path associated matrix to find a tone value aligned with the context as a tone calibration value. The target result of the tonal sequence regulation is in accord with actual situation, extraction of sound tone value is more accurate and reliable, and the process of tonal sequence regulation is controllable. The regulating process can be more flexible by setting related parameters. Good effect can be ensured when the tonal sequence values are widely dispersed, and the robustness is good.

Description

technical field [0001] The invention belongs to the tone extraction and determination part in the audio frequency application technology. The specific content is a method for accurately extracting tones by applying regularization after extracting the original tone sequence. Background technique [0002] Sound is a sound wave generated by the vibration of an object, that is, an analog signal. To process the sound, the computer must digitize the sound wave of the sound, that is, convert it into a digital signal. [0003] An analog signal is a voltage signal or a current signal within a certain range. An analog signal refers to a signal whose information parameters are continuous within a given range, or within a continuous time interval, the characteristic quantity representing information can be presented as Arbitrary numerical signals, so the computer cannot store the analog signal; but according to Shannon's theorem, a continuous function can be completely represented by a...

AI Technical Summary

Problems solved by technology

[0008] The current audio pitch sequence extraction is mainly to obtain the intermediate function waveform consistent with the original audio waveform period through short-term autocorrelation function, short-term average amplitude difference, etc., and then extract the pitch period of the intermediate function waveform; however, in the extraction Due to the various harmonic components of the actual audio and the interference of noise during the process, the pitch extraction value will be uneven, and the exact pitch value in a short period of time cannot be determined;

[0009] The traditional method uses median smoothing algorithm, linear smoothing algorithm or dynamic programming smoothing algorithm to remove outliers in the tone sequence after extracting the tone sequence. These methods are only effective when there are individual mutation points in the stationary sequence; Outliers, after smoothing the values, it is very likely that the outliers cannot be accurately identified, and the anomalies may be enlarged, causing the smoothed sequence to still fluctuate significantly; the method introduced in this invention can solve this problem very well, and the extracted comparison exact pitch value;

Images