EP4047595B1 - Procédé de génération d'un groupe d'échantillons à cohérence de phase pour un instrument virtuel - Google Patents

Procédé de génération d'un groupe d'échantillons à cohérence de phase pour un instrument virtuel Download PDF

Info

Publication number
EP4047595B1
EP4047595B1 EP22153060.3A EP22153060A EP4047595B1 EP 4047595 B1 EP4047595 B1 EP 4047595B1 EP 22153060 A EP22153060 A EP 22153060A EP 4047595 B1 EP4047595 B1 EP 4047595B1
Authority
EP
European Patent Office
Prior art keywords
phase
time
samples
reference sample
phase relationship
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
EP22153060.3A
Other languages
German (de)
English (en)
Other versions
EP4047595C0 (fr
EP4047595A1 (fr
Inventor
Alexander Paprotny
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
E Instruments Lab GmbH
Original Assignee
E Instruments Lab GmbH
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by E Instruments Lab GmbH filed Critical E Instruments Lab GmbH
Publication of EP4047595A1 publication Critical patent/EP4047595A1/fr
Application granted granted Critical
Publication of EP4047595C0 publication Critical patent/EP4047595C0/fr
Publication of EP4047595B1 publication Critical patent/EP4047595B1/fr
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10H—ELECTROPHONIC MUSICAL INSTRUMENTS; INSTRUMENTS IN WHICH THE TONES ARE GENERATED BY ELECTROMECHANICAL MEANS OR ELECTRONIC GENERATORS, OR IN WHICH THE TONES ARE SYNTHESISED FROM A DATA STORE
    • G10H1/00—Details of electrophonic musical instruments
    • G10H1/02—Means for controlling the tone frequencies, e.g. attack or decay; Means for producing special musical effects, e.g. vibratos or glissandos
    • G10H1/06—Circuits for establishing the harmonic content of tones, or other arrangements for changing the tone colour
    • G10H1/12—Circuits for establishing the harmonic content of tones, or other arrangements for changing the tone colour by filtering complex waveforms
    • G10H1/125—Circuits for establishing the harmonic content of tones, or other arrangements for changing the tone colour by filtering complex waveforms using a digital filter
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10H—ELECTROPHONIC MUSICAL INSTRUMENTS; INSTRUMENTS IN WHICH THE TONES ARE GENERATED BY ELECTROMECHANICAL MEANS OR ELECTRONIC GENERATORS, OR IN WHICH THE TONES ARE SYNTHESISED FROM A DATA STORE
    • G10H7/00—Instruments in which the tones are synthesised from a data store, e.g. computer organs
    • G10H7/02—Instruments in which the tones are synthesised from a data store, e.g. computer organs in which amplitudes at successive sample points of a tone waveform are stored in one or more memories
    • G10H7/04—Instruments in which the tones are synthesised from a data store, e.g. computer organs in which amplitudes at successive sample points of a tone waveform are stored in one or more memories in which amplitudes are read at varying rates, e.g. according to pitch
    • G10H7/045—Instruments in which the tones are synthesised from a data store, e.g. computer organs in which amplitudes at successive sample points of a tone waveform are stored in one or more memories in which amplitudes are read at varying rates, e.g. according to pitch using an auxiliary register or set of registers, e.g. a shift-register, in which the amplitudes are transferred before being read
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10H—ELECTROPHONIC MUSICAL INSTRUMENTS; INSTRUMENTS IN WHICH THE TONES ARE GENERATED BY ELECTROMECHANICAL MEANS OR ELECTRONIC GENERATORS, OR IN WHICH THE TONES ARE SYNTHESISED FROM A DATA STORE
    • G10H2250/00—Aspects of algorithms or signal processing methods without intrinsic musical character, yet specifically adapted for or used in electrophonic musical processing
    • G10H2250/541—Details of musical waveform synthesis, i.e. audio waveshape processing from individual wavetable samples, independently of their origin or of the sound they represent
    • G10H2250/615—Waveform editing, i.e. setting or modifying parameters for waveform synthesis

Definitions

  • the present invention relates to phase-coherent samples, their use in a virtual instrument, the virtual instrument, and the method for generating the samples.
  • Playing multiple samples in a virtual instrument is, for example, possible from the US 9, 805, 702 B1 known.
  • the JP 2001100756 A discloses the one-time calculation of an average static phase offset over a temporal portion of the audio tracks on the basis of which a single-channel signal is shifted relative to a single-channel reference signal and thereby the reference signal is made on average more phase-coherent to the single-channel signal.
  • the task is to avoid these disadvantages and yet provide a solution for continuous transitions between samples at different dynamic levels of a real instrument, without having to accept comb filter artifacts or the psychoacoustic impression of two instruments playing in unison, while preserving natural phase fluctuations and time-dynamic changes in the overtone spectrum.
  • the method presented here offers a solution to the problems mentioned above without resorting to resynthesis techniques: By simply Since the playback speed of the original signals is modulated, natural phase and overtone spectrum fluctuations are retained. The procedure is therefore minimally invasive. Multi-channel and spatial signals can be easily converted into a phase-coherent form, while also preserving relative phase relationships, for example between different simultaneous microphone positions in multi-microphone recordings.
  • the method can also be implemented on a standard sequential computer architecture as a real-time processor with moderate computing effort, which enables it to be used, for example, as a plug-in in digital audio processing software or as an on-board effect in a sampler engine.
  • the course of the phase relationship value has in particular at least one change per 100 ms, in particular per 10 ms, and/or the Change of speed and/or time base per 100 ms, in particular per 10 ms, changed at least once.
  • time base refers in particular to the temporal allocation of the individual sample values of the sample and/or audio source.
  • An input audio source is understood to mean in particular an audio source that is processed by the method and/or is used as a reference for the processing of at least one other audio source by the method.
  • the problem is also solved by a method for simultaneously playing at least one sample each of two toner events of at least 500 ms length of the same pitch of a real instrument, wherein the samples are each played back in a phase-coherent manner, in particular for at least 500 ms and at least partially simultaneously.
  • a virtual instrument in particular in a VST, AU or AAX plug-in, having at least one group of phase-coherent samples and/or configured to generate such, wherein the group comprises at least two samples each of a toner event of in particular at least 500 ms in length of the same pitch of a real musical instrument.
  • the virtual instrument is designed to play at least two of the phase-coherent samples of the same pitch at least partially simultaneously and in a phase-coherent manner, whereby the samples have a length of at least 500 ms in particular, in which they do not repeat.
  • At least one sample of the group has a speed change that varies over its time course and/or at least one sample of the group is a processed sampling that has an RMS over its temporal length, in particular without settling and decay times, over the deviations normalized to the mean period in the interval from the mean period of the interval and/or from the period of a reference sample from the group of Samples of less than 75% of the RMS of the unprocessed sample from which the processed sampling was derived, also determined over the same period of time, and/or of the sample corrected by means of a static time offset from which the processed sampling was derived.
  • the varying speed change in particular has at least one change per 100 ms, in particular per 10 ms.
  • the virtual instrument is in particular stored in the form of software and/or a computer program on a data carrier and/or in the form of a computer that is set up as a virtual instrument and/or has such a computer program.
  • the object is also achieved by a system, in particular a hardware and software system, having at least one CPU and a memory, configured to carry out a method according to the invention.
  • a signal processor with at least two inputs for audio signals and at least two outputs for audio signals and configured to sample the signals present at the audio inputs and to carry out the method according to the invention with the samples generated and to output the phase-coherent samples at the outputs.
  • the task is also solved by a group of phase-coherent samples of different dynamics of a pitch of a real musical instrument, whereby the samples have a length of at least 500 ms in which they are phase-coherent and do not repeat.
  • At least one sample of the group has a speed change that varies over its time course and/or at least one sample of the group is a processed sampling that has a speed change that varies over its time length, in particular without attack and decay times, has a RMS determined over the deviations normalized to the mean period in the interval from the mean period of the interval and/or from the period of a reference sample from the group of samples of less than 75% of the RMS also determined over the same period of time of the unprocessed sample from which the processed sampling was derived and/or of the sample corrected by means of a static time offset from which the processed sampling was derived.
  • the varying speed change in particular has at least one change per 100 ms, in particular per 10 ms.
  • a group of samples comprising samples of different dynamics of a pitch of a real musical instrument, wherein the samples each have a length of at least 500 ms, and wherein the group has information on the phase curve of at least one, in particular all samples, over the length of at least 500 ms of the samples, in particular as a track, in particular a MIDI track, in particular an additional track containing the information on the phase curve, in particular one for each sample.
  • a group can be used to easily achieve phase-coherent playback by reducing the differences in the phase curve of the samples during simultaneous playback by modulating the playback speed(s) of the sample(s) when at least two of the samples are played back, at least partially simultaneously.
  • the course of the phase difference can be easily obtained from the information on the phase curve.
  • phase-coherent audio sources in particular audio signals or samples, of at least two different dynamics of a pitch of a real musical instrument also solves the problem, wherein the audio sources have a length of at least 500 ms, in which they are phase-coherent and do not repeat, for the phase-coherent and at least partially simultaneous reproduction of the audio sources of a pitch with at least two different dynamics.
  • a group of audio sources in particular audio signals or samples, of at least two different dynamics of a pitch of a real musical instrument also solves the problem, wherein the audio sources have a length of at least 500 ms, in which they have information on the phase progression, for the phase-coherent and at least partially simultaneous reproduction of the audio sources of a pitch with at least two different dynamics.
  • This is done in particular by determining a course of a phase relationship value, in particular determining a course of the phase deviation, over the duration of the audio source for each of the dependent audio sources and using the course determined in each case to adapt the speed and/or time base of the respective dependent audio source.
  • the phase relationship value changes in particular over the duration of the sample.
  • the special feature of the method according to the invention lies in particular in the realization of a dynamic adaptation by means of which a permanent phase coherence can be achieved even when the phase relationships between the input signals change over time and in which it is also possible to make numerous recordings of the same sound event phase-coherent.
  • the comparison of the phase curve can generally be carried out using a cross-correlator, for example.
  • the passages of the signal, in particular the fundamental frequency, through a predetermined level, in particular the zero crossings can be determined and the position of the zero crossings can be used as information about the temporal position of the edges.
  • the adjustment is carried out in such a way that the offset of the edges, especially zero crossings, is reduced.
  • the course of a phase comparison value is determined, which in particular indicates the deviation of the position of the edges.
  • the course of a phase comparison value is in particular smoothed. This is done in particular by a low-pass filter, the corner frequency of which is set in particular such that no artifacts, in particular no glitches, arise when used to modulate the samples and play the samples at the same time.
  • a look ahead can be used to determine the course of the phase comparison value, which is set in particular after the low-pass filter has been set and/or is set in such a way, in particular increased, that/until no comb filter artifacts occur.
  • the look ahead is preferably in the range of 1 to 50 ms, in particular over 5 ms and/or under 20 ms.
  • phase relationship value also known as a control function
  • a control function which is preferably used to modulate the playback speed and/or to change the speed and/or time base.
  • an average value for modulating the playback speed and/or changing the speed and/or time base which takes into account the phase response of the reference sample and the dependent sample being changed before, at and after the time at which each change is made.
  • the smoothing level is set using a controller via the cut-off frequency of a low-pass filter applied to the control signal.
  • a high cut-off frequency results in a low smoothing level and vice versa.
  • the increasing smoothing level avoids artifacts ("glitches") caused by measurement errors or signal quality, but the correspondingly increased response time also reduces the efficiency of the process and consequently phase-shift-related comb filter effects. can become audible.
  • By setting the look ahead parameter appropriately an effective reaction of the circuit to sudden changes in the phase difference curve can be achieved even with high levels of smoothing.
  • the circuit reacts prematurely and can therefore under certain circumstances create comb filter effects.
  • the parameter is set to a real value between zero and one, whereby with the latter setting the entire buffer time (usually around 60ms) is used for look ahead (i.e. the control signal is applied earlier by the buffer time or the signals to be controlled are delayed by the buffer time), whereas with the former (value of 0) no look ahead takes place at all.
  • the two parameters are preferably set by the user on a signal-specific basis by ear so that a suitable compromise between a short reaction time and freedom from artifacts is achieved with regard to the desired result.
  • a sound engineer is generally familiar with such an approach, since a large number of effects devices common in the industry are operated in a similar way. For example, in a dynamic compressor, the reaction time ("attack” and "release”) and the look-ahead time are adjusted by ear in such a way that a satisfactory compromise between dynamic reduction and freedom from artifacts is achieved.
  • the value of the course of the phase relationship value at a point in time of the reference sample at least partially takes into account the phase course of the reference sample and/or the phase course of another of the audio sources, before, at and/or after the point in time of the reference sample.
  • the time in the reference sample and in the other audio source is in particular identical, in particular counted from their start and/or predetermined times.
  • Such consideration is carried out in particular by a look-ahead and/or a low-pass filter and/or averaging.
  • consideration is given over the entire course, with the exception of the start and end, in particular over 1 to 100 ms at the start and end of the other audio source and/or the reference sample.
  • An audio source can contain several tracks, for example one or more stereo channels.
  • Each audio source is in particular the recording of a real sound event, for example by means of a microphone or a plurality of microphones, in particular at a common position.
  • Static phase shifts due to different run times at different microphone positions, in particular different audio sources, of a common sound event do not represent a disadvantage. They are in particular retained by the method and not eliminated.
  • An audio source is in particular the recording of a sound event, in particular at a position, generated by striking or stroking a string or playing a note on a wind instrument.
  • the dependent audio sources modified in this way form a group of coherent audio sources alone and/or together with the reference sample.
  • One of the audio sources can be used as a reference sample, or an audio signal, in particular with a constant frequency and in particular without a phase jump, in particular a sinusoidal signal.
  • the latter enables time-shifted, phase-coherent playback without having to carry out a phase correction adapted to the time offset, but eliminates the actually desired temporal fluctuations in the frequency of the samples or the phase-coherent audio sources played back.
  • a group or a plurality of audio sources are provided in particular, which can be converted into coherent ones using the method according to the invention. These are stored in particular on a data carrier and/or are in particular part of the system and/or virtual instrument.
  • a first time period, in particular the duration of the transient process, with a length in particular in the range from 0 to 500 ms, in particular in the range from 100 to 400 ms, of the dependent audio sources is not changed in speed and/or time base and a second time period, in particular with a length of more than 500 ms and/or immediately following the first time segment, the or each of the dependent audio sources changes in speed and/or time base.
  • the first and/or second time period is in particular a recording of a real sound event that is not repeated and/or does not repeat.
  • the group of coherent audio sources is in particular designed such that they are coherent when started simultaneously, in particular at least over a duration of at least 500 ms, in which in particular there is no repetition and/or at least after a/the first time period.
  • the group of coherent audio sources is in particular designed such that it comprises or is accompanied by information indicating the temporal sequence in which the partially simultaneous playback is to be started so that the playback is coherent in the area of simultaneous playback.
  • the group of coherent audio sources can also be created depending on settings and/or the time offset of the at least partially simultaneous playback depending on these as needed, in particular before (in particular immediately before and/or a maximum of one hour) and/or during playback. This allows more freedom with regard to the time offset of the playback or its start.
  • the method for playing, the use, the system or the virtual instrument are in particular designed such that the volume of the samples played at least partially simultaneously can be changed relative to one another, in particular varying over the duration of the playback.
  • appropriate setting options are provided in particular.
  • the temporal phase difference curve to the corresponding microphone signal of a dynamic stage to be adjusted is preferably determined by a phase correlator (not necessarily a cross correlator).
  • phase comparison value curve After a phase comparison value curve has been generated, it is preferably processed by suitable unfolding, clipping, filtering and/or Dynamic processors process and/or smooth, for example by means of a low-pass filter with a corner or cut-off frequency in the range of 50 to 200 Hz, in particular from 60 to 150 Hz.
  • An unwrapping method detects the jump points and uses this as a basis to reconstruct the continuous phase difference curve. This can be done, for example, as described in "One-Dimensional Phase Unwrapping Problem" by Dr. Munther Gdeisat and Dr. Francis Lilley, e.g. at https://www.ljmu.ac.uk/-/media/files/ljmu/about-us/faculties-and-schools/fet/geri/onedimensionalphaseunwrapping_finalpdf.pdf.
  • the sequence of the detected zero crossings can be used, since if the edges are determined correctly, a reversal of the phase difference function is immediately apparent from the presence of two consecutive pulses from one of the signals without an intermediate pulse from the other signal.
  • a known unwrapping method can be combined with this method. This makes the unfolding process more precise and less susceptible to interference, which in turn can in many cases enable less invasive smoothing and thus a slightly shorter response time.
  • a clipper can be used to cut off signal peaks beyond a given frequency interval.
  • the lower interval limit of the clipper is preferably identical to the inverse of the buffer time with which the playback of the samples is delayed, the upper limit is preferably fixed, for example set to a value in the range of 10 ⁇ (-7) to 10 ⁇ (-9) Hz.
  • phase relationship value is fed into delay elements as a time parameter, to whose input busses a dependent sample is applied.
  • modulated audio sources that are phase-coherent with the source signal can now be tapped at the outputs of the delay elements.
  • the course of a phase relationship value can be determined for one channel/track only, for several channels/tracks separately, or using a metric that takes several channels/tracks into account.
  • the metric can be an average value, for example.
  • Changing the time base can, for example, be done separately and/or differently for each channel/track of a dependent sample and/or jointly and/or equally for all channels/tracks of a dependent sample.
  • all audio sources and/or all dependent audio sources have the same number of tracks and/or channels.
  • the method can also be used if the number of tracks and/or channels is different.
  • the audio sources are in particular analog or digital recordings of analog acoustic events.
  • all audio sources and/or all dependent audio sources have the same duration.
  • the reference sample has at least the duration of the second longest, in particular the longest, dependent audio sources.
  • reference samples may be recorded by a microphone positioned closer than all of the microphones of the dependent audio sources, in particular very close to the sound source, in particular a maximum of 1 meter from the sound source. This reference sample is then preferably not part of the group of coherent audio sources.
  • the audio sources are preferably recordings of a sound event that is stationary to the microphone(s) used. This means that the musical instrument and microphones used to create the audio sources are preferably stationary.
  • phase relationship determination module for determining the course of the phase deviation. so that the short-term phase fluctuations and deviations characteristic of monophonic musical instrument samples can be quickly compensated.
  • the determination time for the time interval between two edges is preferably essentially given by this, since when an edge arrives, one must wait for the corresponding edge of the second signal.
  • the total reaction time is therefore given in a very good approximation by the absolute temporal phase difference of the given signals plus a value based on the smoothing, in particular their corner or limit frequency, which is preferably in the range of 1 to 20 ms.
  • the reaction time is therefore preferably less than 100 ms, in particular less than 50 ms.
  • a suitable smoothing and/or filtering of the course of the phase deviation before use to change the dependent samples is advantageous, for example by means of a low-pass filter with a corner or cut-off frequency in the range of 50 to 200 Hz, in particular from 60 to 150 Hz.
  • the first harmonic of the fundamental tone can be used temporarily or permanently as a reference instead of the fundamental tone.
  • phase cancellations which in turn are caused by the fact that the first harmonic basically contains no information whatsoever regarding the direction of the corresponding edges in the fundamental, this can lead to new artifacts, in particular now artificially induced fundamental tone failures when the resulting phase-adjusted samples are played synchronously.
  • this can again be easily remedied by a static polarity reversal of the processed reference signal.
  • Polarity reversal (often referred to as "phase rotation" and not entirely correctly) is also a standard Process that is implemented in commercially available mixing consoles and digital audio processing environments.
  • the processed reference signal can be shifted by half a period.
  • the course of a phase relationship value can be determined in various ways, for example analog, digital, in real time or later in time.
  • the fundamental tone of the reference sample and/or the audio sources is preferably first isolated using a steep-edged bandpass filter.
  • the sinusoidal fundamental tone curves are preferably processed by a comparator, resulting in a square wave signal that can then be differentiated.
  • a comparator resulting in a square wave signal that can then be differentiated.
  • both positively and negatively signed edges can also be taken into account to determine the phase phase offset. This leads to a doubling of the resolution of the phase difference determination and can therefore lead to shorter response times, for example within half a period of the signal.
  • the time interval between zero crossings of the reference sample and a dependent audio source is determined and/or used.
  • the staircase function preferably generated from this shows the phase difference curve between the two samples after unwrapping.
  • the course of a phase comparison value is advantageously filtered and/or smoothed.
  • the result is then preferably used as the course of a phase relationship value.
  • the course of a phase relationship value is passed as a control function to the time parameter of a delay element.
  • the change is non-linear processing, in order to avoid aliasing artifacts it is preferably carried out by oversampling the sampling rate of the reference sample by at least four times, in particular at least eight times.
  • the reference sample is provided with additional artificially generated values, in particular amplitude and phase values.
  • This method is used in particular and frequently in commercially available digital emulations of analog signal shapers (such as distortion or overdrive effects) or dynamic processors (such as compressors).
  • each of the adapted audio sources is advantageously band-limited to the original sampling rate or Nyquist frequency, in particular by means of a bandpass filter, and/or sampled with the sampling rate of the digital medium used.
  • the middle C of a flute was recorded one after the other in the dynamic levels piano, mezzoforte and forte with three stereo microphone pairs in the three positions near, medium and far at the same time. There are therefore six channels in total.
  • the left channel of the "near" microphone pair of the mezzoforte dynamic level is selected as the reference sample.
  • the two remaining dependent samples of the piano and forte dynamic levels are now to be adapted to the phase curve of the mezzoforte level.
  • the corresponding channel of the samples of the forte and piano dynamic levels is now compared with the reference sample.
  • the left channel of the "near” microphone position is therefore correlated with the left channel "near” of the reference sample in order to generate the curve of a phase relationship value for each of the dependent samples.
  • Each of the phase relationship values generated by the correlator for each of the dependent samples now controls the time parameter of a delay element, which changes the speed of the respective dependent sample, namely all six microphone channels. This is how the phase-coherent dependent samples are generated.
  • Figure 1 shows schematically and non-limitingly the determination of a course of a phase relationship value (u).
  • the Reference sampling (q) and the dependent sample (s) are first fed to a steep-edged bandpass filter to isolate the fundamental tone.
  • the sinusoidal fundamental tone curves are processed by a comparator, resulting in a square wave, which is then differentiated. All of this does not happen once, but multiple times, in particular continuously and/or at least every 100 ms, in particular at least every 10 ms, over any length of time. This creates a sequence of Dirac pulses which are located exactly at the zero crossings of the samples and are signed positively or negatively depending on the direction of the zero crossing.
  • the subsequent comparator with negative feedback serves to eliminate the negatively signed pulses, which correspond to falling edges.
  • the integrator in the upper part acts as a ramp generator and is restarted by the pulses each time.
  • the signal at its output therefore indicates the time elapsed since the last incoming edge.
  • This is then applied to the input of a sample and hold element (s&h), which is controlled by the pulses from the lower part of the circuit.
  • the output therefore always contains (over the duration of the process) the time difference, the phase relationship value (u), between the edges of the two input signals of the overall circuit.
  • Figure 2 shows schematically and non-limitingly a structure that simultaneously adjusts all microphone signals (so to s N-1 ) of a single dynamic stage, the phase curve of which is to be adjusted to the reference signal (q).
  • the signal is then low-pass filtered and clipped.
  • the clipper which cuts off signal peaks beyond a specified frequency interval, serves the purpose of filtering out peak artifacts.
  • the lower interval limit of the clipper is identical to the inverse buffer time (T buf ), the upper limit in this example is fixed at 10 ⁇ 8ms.
  • phase relationship value also known as the control function (u) (over the duration of the procedure) results.
  • the adjustable buffer time (T buf ) is a basic delay, in particular a constant one, in particular in the range of 5 to 50 ms, for all microphone signals (q and s up to s N-1 ), which is required to enable negative delay times and is added to the control function for the dependent microphone signals.
  • phase-coherent signals ( q , s 0 to s N-1 ).
  • Figure 3 shows schematically and non-restrictively an alternative latency-free implementation.
  • the in the first implementation of the Figure 2 The buffer time required leads to a corresponding latency of the entire circuit.
  • this can be circumvented by using playback modules (metonymically referred to here as "tape") with modulatable playback speed, such as those available in many sampler environments, instead of delay elements.
  • tap playback modules
  • modulatable playback speed such as those available in many sampler environments, instead of delay elements.
  • the RMS norm is chosen, where RMS is normalized by the number of period runs for the purpose of dimensional invariance.
  • the window size was an RMS (root mean square) window according to the Figures 4 to 11 on the horizontal axis in ms.
  • the order of the rows corresponds to the order of the figures.
  • the ratio between the measured phase coherence before and after processing by the procedure or the relative change in the phase coherence can be used as a measure of the quality of the phase alignment procedure.
  • a multi-microphoned sound source was recorded at three different stereo microphone positions and at another single-channel reference microphone position. This resulted in three stereo and one mono tracks.
  • the left channel of a stereo microphone pair (signal 1) was selected as the input audio signal to be processed and the signal from the reference microphone as the reference audio signal (signal 0) and the phase coherence in the sense of the discussed metric between the reference (signal 0) and input audio signal (signal 1) or processed input audio signal (signal 1') was determined.
  • the input audio signal (signal 1) was processed using a constant time shift after the JP 2001100756 A carried out (signal 1").
  • a cello (CE) and a violin (VL) were used as sound sources.
  • Different dynamic levels (f and ff) at different pitches were tested.
  • the sound engineer set the parameters "smoothing level" and "look-ahead” described above by ear with regard to a subjectively optimal ratio between audible phase coherence (i.e., absence of comb filter effects) on the one hand and freedom from artifacts on the other hand, without considering a metric.
  • the Figures 4 to 11 show the progression of the RMS values (RMS values on the vertical axis) with a window of 500 period runs of the measurements on which the table above is based, in the order of the rows in the table.
  • the time in ms is plotted on the horizontal axis. It can be seen here that an improvement can be achieved over large parts of the signal duration.
  • Signal A is signal 1
  • signal B is signal 1"
  • signal C is signal 1'.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Acoustics & Sound (AREA)
  • Multimedia (AREA)
  • General Engineering & Computer Science (AREA)
  • Electrophonic Musical Instruments (AREA)
  • Stereophonic System (AREA)

Claims (6)

  1. Procédé de création d'un groupe de sources audio cohérentes en phase (q, s o à S N-1), en particulier des signaux audio ou des échantillons, comprenant :
    a) Fournir au moins une source audio d'entrée (q, so à S N-1), en particulier des signaux audio ou des échantillons, un premier et un second événement de toner de la même hauteur d'un seul instrument de musique réel,
    b) Sélection d'une source audio, en particulier l'une des sources audio d'entrée du premier événement sonore, comme échantillon de référence (q),
    c) Comparaison de la courbe de phase de l'échantillon de référence (q) avec la courbe de phase d'au moins une des autres sources audio d'entrée (so à S N-1) du second événement sonore sur une première durée de source audio de l'échantillon de référence pour générer une information sur la relation de phase (u),
    d) Modification de la vitesse et/ou de la base de temps d'au moins une source audio d'entrée supplémentaire (so à S N-1) sur une seconde durée de la ou des sources audio d'entrée supplémentaires sur la base de l'information sur la relation de phase (u), caractérisée en ce que l'information sur la relation de phase est un cours d'une valeur de relation de phase variant dans le temps, et que la vitesse et/ou la base de temps varient au cours de la seconde durée en fonction de l'évolution de la valeur de la relation de phase.
  2. Procédé selon la revendication 1, dans lequel la première durée est égale à la seconde durée et/ou la première durée est plus longue que la seconde durée.
  3. Procédé selon l'une quelconque des revendications précédentes, dans lequel la valeur de la valeur de la relation de phase à un moment donné de l'échantillon de référence tient compte, au moins en partie, de l'historique de phase de l'échantillon de référence et/ou de l'historique de phase d'une autre des sources audio, avant, pendant et/ou après le même moment dans l'échantillon de référence, dans lequel les points temporels sont mesurés en particulier par rapport au début de l'événement sonore respectif.
  4. Procédé selon l'une quelconque des revendications précédentes, dans lequel le procédé est mis en oeuvre pour plusieurs seconds événements sonores, en particulier simultanément, avec un premier événement sonore commun et un échantillon de référence commun, et pour chaque second événement sonore, une information de relation de phase est générée et chaque information de relation de phase est utilisée pour modifier la vitesse d'au moins un en particulier plusieurs, source audio d'entrée du second événement sonore respectif.
  5. Procédé selon l'une quelconque des revendications précédentes, dans lequel les étapes (c) et (d) sont réalisées au moyen d'au moins une, en particulier au moyen d'au moins exactement une boucle de commande contre-couplée, dans lequel en particulier une boucle de mesure par second événement sonore est utilisée pour déterminer l'information de relation de phase et/ou une boucle de commande par source audio d'entrée supplémentaire est utilisée pour modifier la vitesse.
  6. Procédé de lecture simultanée d'au moins une source audio, en particulier d'un signal audio ou d'un échantillon, de deux événements sonores, en particulier d'une durée d'au moins 500 ms, de la même hauteur du même instrument réel, dans lequel les sources audio sont lues au moins partiellement simultanément, en particulier sur au moins 500 ms, au moins en phase, et sont générés et/ou sont générés selon la revendication 1.
EP22153060.3A 2021-02-23 2022-01-24 Procédé de génération d'un groupe d'échantillons à cohérence de phase pour un instrument virtuel Active EP4047595B1 (fr)

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
EP21158720 2021-02-23

Publications (3)

Publication Number Publication Date
EP4047595A1 EP4047595A1 (fr) 2022-08-24
EP4047595C0 EP4047595C0 (fr) 2024-08-14
EP4047595B1 true EP4047595B1 (fr) 2024-08-14

Family

ID=74732617

Family Applications (1)

Application Number Title Priority Date Filing Date
EP22153060.3A Active EP4047595B1 (fr) 2021-02-23 2022-01-24 Procédé de génération d'un groupe d'échantillons à cohérence de phase pour un instrument virtuel

Country Status (1)

Country Link
EP (1) EP4047595B1 (fr)

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4779505A (en) * 1983-09-07 1988-10-25 Nippon Gakki Seizo Kabushiki Kaisha Electronic musical instrument of full-wave readout system
DE4008872C2 (de) 1990-03-20 1993-10-28 Wersi Gmbh & Co Verfahren zum Erzeugen von Klängen und elektronisches Musikinstrument
JP3601373B2 (ja) 1999-09-29 2004-12-15 ヤマハ株式会社 波形編集方法
JP6102063B2 (ja) 2011-03-25 2017-03-29 ヤマハ株式会社 ミキシング装置
JP6090204B2 (ja) 2014-02-21 2017-03-08 ヤマハ株式会社 音響信号発生装置
US9805702B1 (en) 2016-05-16 2017-10-31 Apple Inc. Separate isolated and resonance samples for a virtual instrument
US11127387B2 (en) 2016-09-21 2021-09-21 Roland Corporation Sound source for electronic percussion instrument and sound production control method thereof

Also Published As

Publication number Publication date
EP4047595C0 (fr) 2024-08-14
EP4047595A1 (fr) 2022-08-24

Similar Documents

Publication Publication Date Title
DE68913139T2 (de) Störunterdrückung.
DE2818204C2 (de) Signalverarbeitungsanlage zur Ableitung eines störverringerten Ausgangssignals
DE69816221T2 (de) Sprachschnellheitsveränderungsverfahren und vorrichtung
DE4227826C2 (de) Digitales Verarbeitungsgerät für akustische Signale
DE69219718T2 (de) Digitales Datenkodierungs-und Dekodierungsgerät mit hoher Wirksamkeit
DE60103086T2 (de) Verbesserung von quellcodierungssystemen durch adaptive transposition
DE69433073T2 (de) Vorrichtung zür Veränderung akustischer Eigenschaften
DE60024501T2 (de) Verbesserung der perzeptuellen Qualität von SBR (Spektralbandreplikation) UND HFR (Hochfrequenzen-Rekonstruktion) Kodierverfahren mittels adaptivem Addieren von Grundrauschen und Begrenzung der Rauschsubstitution
EP1853089B1 (fr) Méthode pour la suppression de la rétroaction et pour l'éxpansion spéctrale pour des appareils de correction auditive
EP2005421B1 (fr) Dispositif et procédé pour la génération d'un signal d'ambiance
DE602005005186T2 (de) Verfahren und system zur schallquellen-trennung
DE4326746A1 (de) Lautstärkeregelgerät
DE69629934T2 (de) Umgekehrte transform-schmalband/breitband tonsynthese
DE3936693A1 (de) Vorrichtung zum erzeugen digitaler audiosignale
DE102010007842A1 (de) Geräuschunterdrückungsvorrichtung, Geräuschunterdrückungsverfahren und Aufzeichnungsmedium
DE3630692C2 (fr)
DE2720984B2 (fr)
DE69934069T2 (de) Schalleffekt Addiergerät
DE19720651A1 (de) Hörgerät mit verschiedenen Baugruppen zur Aufnahme, Weiterverarbeitung sowie Anpassung eines Schallsignals an das Hörvermögen eines Schwerhörigen
DE3806915A1 (de) Reverb- (nachhall-) generator
DE10009082A1 (de) Gerät zur Wiedergabe einer Wellenform
EP4047595B1 (fr) Procédé de génération d'un groupe d'échantillons à cohérence de phase pour un instrument virtuel
DE3709556C2 (fr)
DE102012204193B4 (de) Audioprozessor und Verfahren zum Verstärken oder Dämpfen eines empfangenen Audiosignals
DE102007011436B4 (de) Vorrichtung und Verfahren zum Formen eines digitalen Audiosignals

Legal Events

Date Code Title Description
PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE APPLICATION HAS BEEN PUBLISHED

AK Designated contracting states

Kind code of ref document: A1

Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: REQUEST FOR EXAMINATION WAS MADE

17P Request for examination filed

Effective date: 20230109

RBV Designated contracting states (corrected)

Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

RIC1 Information provided on ipc code assigned before grant

Ipc: G10H 1/12 20060101ALI20240201BHEP

Ipc: G10H 7/04 20060101ALI20240201BHEP

Ipc: G10H 7/00 20060101AFI20240201BHEP

GRAP Despatch of communication of intention to grant a patent

Free format text: ORIGINAL CODE: EPIDOSNIGR1

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: GRANT OF PATENT IS INTENDED

INTG Intention to grant announced

Effective date: 20240319

GRAS Grant fee paid

Free format text: ORIGINAL CODE: EPIDOSNIGR3

GRAA (expected) grant

Free format text: ORIGINAL CODE: 0009210

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE PATENT HAS BEEN GRANTED

AK Designated contracting states

Kind code of ref document: B1

Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

REG Reference to a national code

Ref country code: GB

Ref legal event code: FG4D

Free format text: NOT ENGLISH

REG Reference to a national code

Ref country code: CH

Ref legal event code: EP

REG Reference to a national code

Ref country code: DE

Ref legal event code: R096

Ref document number: 502022001422

Country of ref document: DE

REG Reference to a national code

Ref country code: IE

Ref legal event code: FG4D

Free format text: LANGUAGE OF EP DOCUMENT: GERMAN

U01 Request for unitary effect filed

Effective date: 20240823

U07 Unitary effect registered

Designated state(s): AT BE BG DE DK EE FI FR IT LT LU LV MT NL PT RO SE SI

Effective date: 20240903

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: NO

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20241114

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: PL

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20240814

Ref country code: GR

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20241115

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: IS

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20241214

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: HR

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20240814

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: RS

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20241114

Ref country code: ES

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20240814

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: RS

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20241114

Ref country code: PL

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20240814

Ref country code: NO

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20241114

Ref country code: IS

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20241214

Ref country code: HR

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20240814

Ref country code: GR

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20241115

Ref country code: ES

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20240814

U20 Renewal fee for the european patent with unitary effect paid

Year of fee payment: 4

Effective date: 20250127

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: SM

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20240814

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: CZ

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20240814

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: SK

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20240814

PLBE No opposition filed within time limit

Free format text: ORIGINAL CODE: 0009261

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: NO OPPOSITION FILED WITHIN TIME LIMIT

26N No opposition filed

Effective date: 20250515

REG Reference to a national code

Ref country code: CH

Ref legal event code: PL

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: MC

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20240814

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: CH

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20250131

U20 Renewal fee for the european patent with unitary effect paid

Year of fee payment: 5

Effective date: 20251201

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: IE

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20250124

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: GB

Payment date: 20260122

Year of fee payment: 5