Signal processing method and sound source data forming apparatus

Claim 1

A signal recording method comprising; supplying an input analog signal or an input digital signal corresponding to the analog signal through a comb filter having only fundamental frequency band of the input analog signal and frequency bands of harmonic components thereof as pass bands to produce output analog or digital signals; and extracting looping domains of the output analog or digital signals and recording the looping domains on a storage medium.

Claim 3

A method for detecting a pitch of an analog signal comprising: sampling the analog signal to convert it to a digital signal; subjecting the digital signal produced upon the digital conversion of the analog signal to a Fourier transform to produce a plurality of frequency components; matching the phases of the frequency components; subjecting the frequency components to an inverse Fourier transform to produce output data; and detecting the periods of the peaks of the output data to thereby detect the pitch of the analog signal.

Claim 4

A method for detecting a pitch of an analog signal comprising: sampling the analog signal to convert it to a digital signal; subjecting the digital signal produced upon the digital conversion of the analog signal to a Fourier transform to produce a plurality of frequency components; matching the phases of the frequency components; subjecting the frequency components to an inverse Fourier transform to produce output data; and detecting the periods of the peaks of the output data to thereby detect the pitch of the analog signal.

Claim 5

A method of compressing an input signal containing blocks of digital data comprising: providing a plurality of modes consisting of directly outputting the input signal and outputting the input signal through a filter; selecting for each block of the input signal the one of the plurality of modes which will produce an output signal having the highest compression rate; and prior to a first block of the input signal, affixing to the input signal a block which contains a simulated input signal which will cause the mode of directly outputting the input signal to be selected for the first block of the input signal; and processing the input signal inclusive of the simulated input signal.

Claim 6

A method of compressing an input signal containing blocks of digital data comprising: providing a plurality of modes consisting of directly outputting the input signal and outputting the input signal through a filter; selecting for each block of the input signal the one of the plurality of modes which will produce an output signal having the highest compression rate; and prior to a first block of the input signal, affixing to the input signal a block which contains a simulated input signal which will cause the mode of directly outputting the input signal to be selected for the first block of the input signal; and processing the input signal inclusive of the simulated input signal.

Claim 7

A method of processing digital waveform data containing looping domains constituted of an integral number n of constant periods of waveform data comprising; dividing each of the constant periods of waveform data into an integral number of blocks, each block consisting of an integral number of words and each constant period of waveform data containing at least a starting block and an end block; providing a plurality of modes for processing the data, including directly outputting the data and filtering the data on either a first or second order differential; selecting for each block of data the one of the plurality of processing modes which results in the most compression of the data; and forming bit compression and encoding parameters for the starting block on the basis of data of the starting block and the end block.

Claim 8

A method of processing digital waveform data containing looping domains constituted of an integral number n of constant periods of waveform data comprising; dividing each of the constant periods of waveform data into an integral number of blocks, each block consisting of an integral number of words and each constant period of waveform data containing at least a starting block and an end block; providing a plurality of modes for processing the data, including directly outputting the data and filtering the data on either a first or second order differential; selecting for each block of data the one of the plurality of processing modes which results in the most compression of the data; and forming bit compression and encoding parameters for the starting block on the basis of data of the starting block and the end block.

Claim 11

A method for producing sound source data for storage in a storage medium by processing an input musical sound signal having repetitive waveforms, comprising; sampling the analog input musical sound signal to convert it to a digital input signal; detecting a pitch which is a fundamental frequency of the input musical sound signal by subjecting the digital input signal to a Fourier transform to produce a plurality of frequency components, matching phases of the frequency components, subjecting the frequency components to an inverse Fourier transform to produce output data, and detecting periods of the peaks of the output data to thereby detect the pitch of the analog signal; comb filtering the sampled analog input musical sound signal and outputting a comb filtered signal, the comb filtering having only a frequency band of the fundamental frequency detected by the pitch detection means and a frequency band of harmonic components thereof as a pass band; generating looping data by finding values of predetermined evaluation functions at a plurality of sets of the samples of the comb filtered signal, each set having two points spaced relatively from each other by a repetitive period corresponding to the detected pitch of the digital input signal and extracting as repetitive data plural samples between the two points the evaluation functions of which have values indicating a high similarity of the waveforms in a vicinity of the two points; producing waveform data by mixing the looping data with a simulated digital input signal occurring during the period proceeding the start point of the analog input musical sound signal; and compressing and encoding an integral number n of constant periods of the waveform data into compressed data words and parameters concerned with compression into an integral number m blocks of data each containing an integral number h of samples, each block containing at least a start block and an end block, and the parameters for the start block being formed on the basis of the data for the start block and the end block.

Claim 12

A method for producing sound source data for storage in a storage medium by processing an input musical sound signal having repetitive waveforms, comprising; sampling the analog input musical sound signal to convert it to a digital input signal; detecting a pitch which is a fundamental frequency of the input musical sound signal by subjecting the digital input signal to a Fourier transform to produce a plurality of frequency components, matching phases of the frequency components, subjecting the frequency components to an inverse Fourier transform to produce output data, and detecting periods of the peaks of the output data to thereby detect the pitch of the analog signal; comb filtering the sampled analog input musical sound signal and outputting a comb filtered signal, the comb filtering having only a frequency band of the fundamental frequency detected by the pitch detection means and a frequency band of harmonic components thereof as a pass band; generating looping data by finding values of predetermined evaluation functions at a plurality of sets of the samples of the comb filtered signal, each set having two points spaced relatively from each other by a repetitive period corresponding to the detected pitch of the digital input signal and extracting as repetitive data plural samples between the two points the evaluation functions of which have values indicating a high similarity of the waveforms in a vicinity of the two points; producing waveform data by mixing the looping data with a simulated digital input signal occurring during the period proceeding the start point of the analog input musical sound signal; and compressing and encoding an integral number n of constant periods of the waveform data into compressed data words and parameters concerned with compression into an integral number m blocks of data each containing an integral number h of samples, each block containing at least a start block and an end block, and the parameters for the start block being formed on the basis of the data for the start block and the end block.

Claim 13

A method for detecting a pitch of an analog signal comprising: electronically sampling the analog signal to convert it to a digital signal; Fourier transforming the digital signal produced upon the digital conversion of the analog signal to produce a plurality of frequency components; electronically matching the phases of the frequency components; inverse Fourier transforming the frequency components to produce output data; and electronically detecting the periods of the peaks of the output data to thereby detect the pitch of the analog signal and outputting a signal representative of the detected pitch.

Claim 14

A method for detecting a pitch of an analog signal comprising: electronically sampling the analog signal to convert it to a digital signal; Fourier transforming the digital signal produced upon the digital conversion of the analog signal to produce a plurality of frequency components; electronically matching the phases of the frequency components; inverse Fourier transforming the frequency components to produce output data; and electronically detecting the periods of the peaks of the output data to thereby detect the pitch of the analog signal and outputting a signal representative of the detected pitch.