EP1653443A1 - Synthétiseur de son polyphonique - Google Patents

Synthétiseur de son polyphonique Download PDF

Info

Publication number
EP1653443A1
EP1653443A1 EP04077975A EP04077975A EP1653443A1 EP 1653443 A1 EP1653443 A1 EP 1653443A1 EP 04077975 A EP04077975 A EP 04077975A EP 04077975 A EP04077975 A EP 04077975A EP 1653443 A1 EP1653443 A1 EP 1653443A1
Authority
EP
European Patent Office
Prior art keywords
sound
sounds
storage element
transform
processing parameters
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.)
Withdrawn
Application number
EP04077975A
Other languages
German (de)
English (en)
Inventor
Conor Mcnally
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.)
Silicon Ip Ltd
Original Assignee
Silicon Ip Ltd
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 Silicon Ip Ltd filed Critical Silicon Ip Ltd
Priority to EP04077975A priority Critical patent/EP1653443A1/fr
Publication of EP1653443A1 publication Critical patent/EP1653443A1/fr
Withdrawn legal-status Critical Current

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
    • G10H7/00—Instruments in which the tones are synthesised from a data store, e.g. computer organs
    • G10H7/08—Instruments in which the tones are synthesised from a data store, e.g. computer organs by calculating functions or polynomial approximations to evaluate amplitudes at successive sample points of a tone waveform
    • G10H7/10—Instruments in which the tones are synthesised from a data store, e.g. computer organs by calculating functions or polynomial approximations to evaluate amplitudes at successive sample points of a tone waveform using coefficients or parameters stored in a memory, e.g. Fourier coefficients
    • 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
    • G10H2250/131—Mathematical functions for musical analysis, processing, synthesis or composition
    • G10H2250/215—Transforms, i.e. mathematical transforms into domains appropriate for musical signal processing, coding or compression
    • G10H2250/235—Fourier transform; Discrete Fourier Transform [DFT]; Fast Fourier Transform [FFT]
    • 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
    • G10H2250/131—Mathematical functions for musical analysis, processing, synthesis or composition
    • G10H2250/261—Window, i.e. apodization function or tapering function amounting to the selection and appropriate weighting of a group of samples in a digital signal within some chosen time interval, outside of which it is zero valued
    • G10H2250/285—Hann or Hanning window
    • 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
    • G10H2250/645—Waveform scaling, i.e. amplitude value normalisation

Definitions

  • the invention relates to a polyphonic synthesizer and a method of synthesizing multiple sounds of multiple timbres.
  • the invention provides a polyphonic synthesiser capable of implementing all synthesis of multiple sounds efficiently in the frequency domain.
  • polyphonic synthesisers are available in many forms, from your mobile phone ring-tone player to a professional keyboard synthesizer.
  • Most polyphonic Synthesizers use the MIDI protocol, http:// www.midi.org , as a means of representing and reading a musical composition.
  • Most polyphonic synthesizers use time domain FM or Wavetable synthesis (or a combination of both) as the main synthesis technique.
  • FM synthesis is a method of approximating an instrument's spectral content but in general it will sound synthesized or unrealistic.
  • Wavetable synthesizers store samples of real instruments and then play these samples back during synthesis to give a more realistic sound.
  • a basic requirement for a polyphonic synthesizer is a Soundbank that generally stores information required by the synthesizer to reproduce the instruments or sounds it is required to synthesize.
  • the synthesizer should also be capable of synthesizing multiple sounds of multiple timbres simultaneously.
  • the number of sounds that are played at any one time denotes the degree of polyphony.
  • the volume and frequency of each of the sounds should be adjustable at any time during the playback of the sounds.
  • the synthesizer should be able to respond to 'Note On' and 'Note Off' commands in a way that relates to the instrument that is being played.
  • a phase vocoder approach may be used in synthesising sounds.
  • the phase vocoder normally utilises a known technique that uses frequency-domain transformations to implement a variety of audio effects such as time and pitch scaling of sounds. These known techniques are described in 'The Phase Vocoder: Theory and Practice' Organised Sound 2, Cambridge University Press, 1997, Fischman R.
  • the Phase Vocoder generally uses a method of pitch scale modification to the notes or sounds through a combination of time scaling and sample rate conversion to synthesize the notes.
  • a typical method of modifying the pitch scale of a note is described in 'Improved Phase Vocoder Time Scale Modification of Audio', IEEE, LaRoche J, Vol. 7, No. 3, May 1999.
  • IFFT inverse Fast Fourier Transform
  • Patent number US 5,686,683 B1 entitled 'Inverse Transform Narrowband / Broadband Sound Synthesizer' discloses a variety of broadband features to synthesise a more complex sound. These features include numerous different frequency domain noise generation, filtering, interpolation techniques that can be used to generate a more complex sound.
  • the main problem with the operation of this US patent is that each Voice (i.e. an instrument) of a polyphonic sound requires an individual inverse Fourier Transform.
  • the other Voices in the Polyphonic sound require an additional IFFT for each voice.
  • the output of each IFFT for each voice is then blended in time to give the overall polyphonic sound.
  • This multiple transform operation application is computationally very intensive and technically difficult to achieve. The application of synthesizing the notes is thus restricted by the multiple transform operation.
  • One of the primary advantages of the present invention is that when all the sounds are processed individually and accumulated in the storage element or a buffer for a particular time frame a single transform is only required to be executed on all of the accumulated processed sounds for a particular time frame in the buffer to provide a real value output to synthesise multiple sounds.
  • the single transform operation is a single inverse Fast Fourier Transform (IFFT).
  • IFFT inverse Fast Fourier Transform
  • the step of specifying each sound in terms of a frequency domain representation consisting of a series of time overlapped frames in combination with said set of processing parameters relating to the characteristics of at least one sound is carried out.
  • An advantage of using the frequency domain is that before executing said inverse transform operation the invention can carry out additional processing on the sounds in the frequency domain representation.
  • the additional processing on the sounds comprises one or more of the following: reverberation, headphone externalisation, positional audio, environmental modelling, stereo widening, equalisation and/or bass boost. This has the advantage that the invention can perform additional processing extremely efficiently and with higher quality sound synthesis.
  • the invention carries out the steps of reconstructing the time sampled representation of the sound by:
  • the step of pitch scaling at least one sound stored in a sound bank to any sound in a key range associated with at least one timbre is carried out.
  • the step of frequency scaling at least one sound stored in said sound bank to a different sound frequency is carried out.
  • Frequency scaling at least one sound stored in said sound bank to a different sound frequency may be carried out to implement at least one of the following: pitch bend, legato and/or portamento.
  • the step of calculating an envelope scale factor or scale factors on a per frame basis and applying at least one scale factor to at least one processing parameter to generate an envelope scaling factor or scaling factors for a frame is carried out.
  • the invention carries out the step of combining the volume of each sound for a frame with the envelope scale factor or scale factors to provide a total envelope scale factor.
  • the processing of a plurality of sounds is carried out in parallel with at least one of said applied processing parameters.
  • the invention carries out the step of sharing the processing during a time frame for simultaneous sounds of different frequency for the same timbre.
  • the accumulated processed information for all sounds for each frame may be stored in frequency bins in said storage element and a scaling factor may be applied to each bin before said single transform is executed.
  • accumulated processed information for all sounds for each frame is stored in frequency bins in said storage element and a scaling factor may be applied to each bin during execution of said single transform.
  • the scaling value can be determined from an absolute sum of the accumulated information in said storage element.
  • the step of applying reverse scaling to said transform output for each time frame, applying a windowing function and overlapping and adding the resulting time frame with at least one preceding time frame is executed.
  • a computer and/or a computer program comprising program instructions for causing a computer to carry out the above method which may be embodied on a record medium, carrier signal or read-only memory.
  • a sound bank 2 can generate a sound bank of sounds to store at least one note per timbre.
  • a "sound” is meant to encompass any physical sound whether it be a musical note or a sound effect of any kind.
  • the different sounds that a synthesiser can produce are sometimes called “patches”, “programs”, “algorithms” or “timbres”.
  • Programmable synthesisers commonly assign "program numbers” or "patch numbers” to each sound. For instance, a sound module might use patch number one for its acoustic piano sound, and patch number thirty six for its fretless bass sound.
  • a synthesiser can generally be regarded as multi-timbral if it can generate two or more different instrument sounds simultaneously. The same sound or note can be played on two different instruments and may sound completely different since each instrument has its own timbre. For the sake of clarity, the present description uses the terms sound and timbre in conjunction with the operation of the synthesizer, but should not be limited to these terms.
  • the sound bank 2 is in communication with a channel block 3, which drives the synthesis process, and a polyphonic synthesiser 4.
  • the sound information stored in the sound bank 2 can be generated using a vocoder analysis technique.
  • the polyphonic synthesiser 4 outputs synthesised sounds to a storage element 5, for example a buffer or a memory store.
  • the system architecture also includes an overlap and add (OAA) block 6, a timing block 7, memory blocks 8, 9 & 10 and a file parser 11 the operation of each will be explained in more detail below in the description.
  • OAA overlap and add
  • the sound bank 2 contains a frequency analysed timbre set to support timbres of a specified standard, for example the standard set in the Midi specification, http://www.midi.org.
  • the File parser 11 decodes the score sheet which could be in the form of a MIDI file, to be synthesized by the channel block 3 and synthesiser 4.
  • the file parser 11 outputs timed events detailing the sequence of instruments/notes/parameters to be used in synthesis.
  • the Channel block 3, which drives the synthesis process can include a processor. Input events from the file parser 11, the timing block 7 and soundbank 2 trigger queuing and the setting up of the necessary parameters and modules needed to synthesize the requested sounds.
  • the input events can be controlled directly by a user or input from a pre-programmed application or content file.
  • the synthesizer 4 comprises sixty four Synthesizer modules to support sixty four note polyphony.
  • Each synthesizer module is assigned sounds or notes to synthesize from one of a plurality of Channel Modules in the channel block 3.
  • Each synthesizer will synthesize the sounds on a per frame basis, the length of a frame is determined by the timing module 7.
  • Each sound is specified in terms of a frequency domain representation comprising of a series of time overlapped frames in combination with a set of processing parameters relating to the characteristics of at least one sound.
  • the output frames, which are the processed information of each sound, from each synthesizer are accumulated into the storage element or buffer 5 for an individual frame.
  • the buffer 5 handles frequency to time domain conversion by executing a single inverse Fast Fourier Transform (IFFT) on all the accumulated information.
  • IFFT inverse Fast Fourier Transform
  • the IFFT operates on all the frequency information accumulated into the buffer 5 using a single transformation to provide a real output.
  • the overlap and add (OAA) block 6 uses the real output of the IFFT to reconstruct the synthesized waveform.
  • the Channel block 3 includes a Channel Driver 31, Channel Synthesizer Assignment 32, Channel Buffers 33, Instrument 34 and Remove Note block 35.
  • the operation of the Channel Block 3 is common for both Melodic and Percussion instruments or timbres or sound effects.
  • An example of the difference between the two timbres in the MIDI standard is that Percussion instruments events are only received on MIDI Channel 9 and Melodic instruments are received on all other 15 MIDI channels (i.e. 0-8 &10 -15).
  • the Soundbank 2 is accessed with the requested program/sound number to return the Soundbank memory location for the requested instrument 34 to the channel block 4.
  • the requested instrument 34 can be mapped to a different timbre internally in the Soundbank 2. This is used to reduce the number of timbres stored in the Soundbank 2.
  • the corresponding source note key range for each instrument and memory location of each stored sound is extracted from the Soundbank 2.
  • Instrument 34 also calls a Header Info block to extract header information of each source sound.
  • Each source sound or note has a header section, which contains the configuration parameters used for processing that are critical to drive the synthesis of that sound.
  • the configuration parameters can include:
  • the Channel buffers 33 store the requested Note On numbers, sound velocities and source sound index values assigned by the Channel Driver 31.
  • the Remove Note block 35 handles Note Off events by removing the corresponding note information from the channel buffers and from the assigned synthesizer.
  • the Channel Synth Assign block 32 assigns the sounds set-up and queued in the Channel Buffers 33 to synthesizers within the synthesizer block 4. Up to four sounds that are frequency scaled from the same source sound can be grouped into a single synthesizer module. Additional sounds for the same source sound are assigned to different synthesizers in groups of four.
  • the synthesizer 4 is loaded with Channel number (i.e. instrument pointer) and a pointer to the source sound header information (i.e. which sound to frequency scale from).
  • the synthesizer 4 is loaded with the sound numbers, sound velocities and frequency scaling values to synthesize.
  • the synthesizer 4 will synthesize a new frame for each of the sounds that are assigned to it in the frequency domain representation and accumulate each sound frame into the storage element 5.
  • Each synthesizer 40a-d can synthesize up to four sounds from the same source sounds, for example a chord of 4 simultaneous Piano sounds may be frequency scaled from the same source sound. Sounds are assigned to synthesizer modules 40a-d in the Channel block 3.
  • the synthesizer 4 is loaded with information of sound number, velocity, frequency scale value, frame count number, source note header pointer information.
  • the synthesizer 4 incorporates the main synthesizer functions block, a memory buffer 41, a memory access controller 42 and an Envelope generator 43.
  • the Envelope generator 43 calculates the Envelope magnitude on a per frame basis and is combined with sound volumes to give the total sound Envelope magnitude. If the Envelope magnitude has decayed to zero there is no need to synthesize a frame for the currently selected synthesizer. Otherwise a new frame is synthesized for the sounds used by the current synthesizer.
  • the memory access controller is used to update the memory pointers used to read Bin, Mag, and Phase values from the Soundbank 2 for the current frame.
  • the basic synthesis procedure can be optimised for various different synthesis requirements. In some applications no frequency scaling is required as is the case with percussion instruments.
  • a synthesis algorithm is broken out into separate functions, each facilitating an optimised synthesis path for each of the possible synthesis scenarios. Each synthesis function is a variation of the same synthesis algorithm.
  • the Envelope Generator block 43 handles generation of the waveform Envelope during synthesis.
  • An Envelope scale factor is calculated on a per frame basis and applied to the note velocity values, assigned to the synthesizer by the Channel 3, to give the total Envelope scaling factor.
  • the Envelope generator 43 employs a polynomial generator 431 capable of generating linear, quadratic or cubic curves and Envelope generation control 432.
  • the polynomial coefficients (P o , P 2 , P 2 , P 3 ) are read by the Channel 3 on each new program change command.
  • Channel 3 invokes initialisation of the new polynomial coefficients when a program change is received.
  • the polynomial is calculated on a per frame basis.
  • the Memory Access controller 42 handles control of the Soundbank 2 memory location pointers. These pointers point to the Soundbank memory locations that correspond to the position where the Bin, Magnitude and Phase values for the current frame are stored.
  • the Memory Access control 42 pointers are updated on a per frame basis.
  • the Attack phase is the initial striking of the sound where the most transients occur and the most rich dynamics of the sound occur.
  • the Decay phase immediately follows the Attack phase where the sound reduces in amplitude to a steady state called the Sustain phase.
  • the Sustain phase is generally simpler in terms of frequency content and decays gently over a period of time. This phase lends itself well to looping frames and just applying a decaying envelope in order to save on memory. Finally, when the sound is released, it decays to zero over a period of time.
  • Each of these phases of the envelope can be controlled independently in terms of their duration and shape.
  • Other envelopes could also be used but the ADSR or variations are generally accepted to be the best for musical sounds, for example DAHDSR (Delay, Attack, Hold, Decay, Sustain, Release) envelope.
  • the storage element 5 handles storage and conversion of the frequency-domain synthesis frames into Time domain synthesis frames.
  • the storage element 5 comprises Real and Imaginary IFFT buffers 51 and 52, an IFFT scaling block 53 and the IFFT routine 54.
  • the IFFT buffers 51 and 52 comprise of two (Real and Imaginary) 1024x2byte arrays which act as a transformer to execute the inverse transform.
  • the lower half (513 Bins) of the buffers are used.
  • the lower 513 bins of the IFFT buffers are reset to zero.
  • the processed bin information of each note frame is accumulated into the IFFT buffer 51 and 52.
  • the buffers are scaled by the scaling block 53 to avoid overflow in the IFFT calculation.
  • the scaling factor is determined by using the result of the sum of the absolute value of the IFFT buffers to drive threshold detectors, if the sum is determined to exceed a certain threshold.
  • the scaling factor can be applied with one of two methods:
  • the present invention can support both options but the first is preferable for optimising computational requirements as the improvement in sound quality offered by the second is not perceivable. This is mainly due to the fact that scaling is only used during loud (high polyphony) sections of the synthesized sound and small quantization errors during a loud section are not perceivable. The effect of scaling is then removed from the output of the single IFFT calculation. This is done in the OAA block 7. It will be appreciated that the use of a single transform only, improves the synthesising process as only a single calculation is needed.
  • the IFFT routine is a standard 1024-point (16Bit) IFFT.
  • the IFFT does not require any overflow checking/scaling support, although IFFT routines that incorporate such features may also be utilized.
  • IFFT routines that incorporate such features may also be utilized.
  • During synthesis only the bins up to N/2 +1 (i.e. 513) are used. This is because the FFT has conjugate symmetry about the point N/2 +1.
  • Some IFFT routines copy the conjugate information internally and some do not.
  • the overlap and add (OAA) block 6 uses the real output of the buffer 5 to reconstruct the synthesized waveform.
  • the OAA block 6 contains scaling and windowing functions 61 and an OAA buffer 62.
  • the scaling that was applied to the input of the IFFT is removed from the real output from the IFFT.
  • a Hanning window from the memory block 8 is first applied to the IFFT frame.
  • the Hanning window is stored as a look up table in the memory block 8. After scaling and window tapering the resulting frame is overlapped and added to the preceding OAA frames, where the overlap is equal to Hop samples (i.e.256).
  • Hop samples i.e.256
  • the Polyphonic Synthesizer described in this specification can be used as a midi player, however the invention is not restricted to the midi specification and can be used for general polyphonic sound synthesis. It will be appreciated that the application is not limited to be applied to the Midi player standard and can be used in any other synthesizing format, for example sound formats besides MIDI include: SP-MIDI, DLS (Downloadable Sounds), SMAF, iMelody and XMF (eXtensible Music Format).
  • a polyphonic synthesizer/midi-player that is capable of implementing all synthesis in the frequency domain, with use of the Phase Vocoder, can be applied to the present invention.
  • the term polyphonic refers to the ability of a synthesiser to play more than one sound at a time.
  • storage element and buffer are used interchangeably and should be interpreted broadly.
  • the embodiments in the invention described with reference to the drawings comprise a computer apparatus and/or processes performed in a computer apparatus.
  • the invention also extends to computer programs, particularly computer programs stored on or in a carrier adapted to bring the invention into practice.
  • the program may be in the form of source code, object code, or a code intermediate source and object code, such as in partially compiled form or in any other form suitable for use in the implementation of the method according to the invention.
  • the carrier may comprise a storage medium such as ROM, e.g. CD ROM, or magnetic recording medium, e.g. a floppy disk or hard disk.
  • the carrier may be an electrical or optical signal which may be transmitted via an electrical or an optical cable or by radio or other means.

Landscapes

  • Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Engineering & Computer Science (AREA)
  • Algebra (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Pure & Applied Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Acoustics & Sound (AREA)
  • Multimedia (AREA)
  • Electrophonic Musical Instruments (AREA)
EP04077975A 2004-10-29 2004-10-29 Synthétiseur de son polyphonique Withdrawn EP1653443A1 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
EP04077975A EP1653443A1 (fr) 2004-10-29 2004-10-29 Synthétiseur de son polyphonique

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
EP04077975A EP1653443A1 (fr) 2004-10-29 2004-10-29 Synthétiseur de son polyphonique

Publications (1)

Publication Number Publication Date
EP1653443A1 true EP1653443A1 (fr) 2006-05-03

Family

ID=34928616

Family Applications (1)

Application Number Title Priority Date Filing Date
EP04077975A Withdrawn EP1653443A1 (fr) 2004-10-29 2004-10-29 Synthétiseur de son polyphonique

Country Status (1)

Country Link
EP (1) EP1653443A1 (fr)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5401897A (en) * 1991-07-26 1995-03-28 France Telecom Sound synthesis process
US5686683A (en) * 1995-10-23 1997-11-11 The Regents Of The University Of California Inverse transform narrow band/broad band sound synthesis
WO2000007114A1 (fr) * 1998-07-27 2000-02-10 Koninklijke Philips Electronics N.V. Processeur de tfr et procede de prevention des surcharges
US6137839A (en) * 1996-05-09 2000-10-24 Texas Instruments Incorporated Variable scaling of 16-bit fixed point fast fourier forward and inverse transforms to improve precision for implementation of discrete multitone for asymmetric digital subscriber loops

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5401897A (en) * 1991-07-26 1995-03-28 France Telecom Sound synthesis process
US5686683A (en) * 1995-10-23 1997-11-11 The Regents Of The University Of California Inverse transform narrow band/broad band sound synthesis
US6137839A (en) * 1996-05-09 2000-10-24 Texas Instruments Incorporated Variable scaling of 16-bit fixed point fast fourier forward and inverse transforms to improve precision for implementation of discrete multitone for asymmetric digital subscriber loops
WO2000007114A1 (fr) * 1998-07-27 2000-02-10 Koninklijke Philips Electronics N.V. Processeur de tfr et procede de prevention des surcharges

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
DATABASE WPI Derwent World Patents Index; AN 2000-183286 *
DATABASE WPI Derwent World Patents Index; AN 2001-101357 *
GEORGE B ET AL: "Fixed frame rate ABS/OLA sinusoidal modeling applied to polyphonic music synthesis", MULTIMEDIA SIGNAL PROCESSING, 1997., IEEE FIRST WORKSHOP ON PRINCETON, NJ, USA 23-25 JUNE 1997, NEW YORK, NY, USA,IEEE, US, 23 June 1997 (1997-06-23), pages 59 - 64, XP010233897, ISBN: 0-7803-3780-8 *
X. RODET AND P. DEPALLE: "SPECTRAL ENVELOPES AND INVERSE FFT SYNTHESIS", JOURNAL OF THE AUDIO ENGINEERING SOCIETY, 4 October 1992 (1992-10-04), XP002313030 *

Similar Documents

Publication Publication Date Title
EP1688912B1 (fr) Synthétiseur vocal de sons multiples
US6182042B1 (en) Sound modification employing spectral warping techniques
US6687674B2 (en) Waveform forming device and method
US6881888B2 (en) Waveform production method and apparatus using shot-tone-related rendition style waveform
JP4181637B2 (ja) ウェーブテーブルシンセサイザに使用される音響サンプルを予め処理するための期間強制フィルタ
WO1997017692A9 (fr) Synthetiseur musical a modelisation parametrique des signaux
EP1039442B1 (fr) Méthode et dispositif pour la compression et la génération d'une forme d'onde
JP2008112183A (ja) 音響シンセサイザにおける低減メモリリバーブシミュレータ
US7396992B2 (en) Tone synthesis apparatus and method
JP4179268B2 (ja) データ合成装置およびデータ合成処理のプログラム
US4205577A (en) Implementation of multiple voices in an electronic musical instrument
JP3775319B2 (ja) 音楽波形のタイムストレッチ装置および方法
EP1653443A1 (fr) Synthétiseur de son polyphonique
JP2010217475A (ja) 楽音信号発生装置
JPH08160961A (ja) 音源装置
Derenyi et al. Synthesizing trumpet performances
JPH02187796A (ja) 実時間ディジタル加算シンセサイザ
JP3404850B2 (ja) 音源装置
JP3719129B2 (ja) 楽音信号合成方法、楽音信号合成装置および記録媒体
JP5493408B2 (ja) 波形データ生成方法
JP3613191B2 (ja) 波形生成方法及び装置
JP3788096B2 (ja) 波形圧縮方法及び波形生成方法
JP3802293B2 (ja) 楽音処理装置および楽音処理方法
JP2734024B2 (ja) 電子楽器
JP5935346B2 (ja) 楽音合成装置及びプログラム

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

AK Designated contracting states

Kind code of ref document: A1

Designated state(s): AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HU IE IT LI LU MC NL PL PT RO SE SI SK TR

AX Request for extension of the european patent

Extension state: AL HR LT LV MK

AKX Designation fees paid
REG Reference to a national code

Ref country code: DE

Ref legal event code: 8566

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

Free format text: STATUS: THE APPLICATION IS DEEMED TO BE WITHDRAWN

18D Application deemed to be withdrawn

Effective date: 20061104