Recognition or synthesis of overtones produced by human speech

JP2024518630A5Active Publication Date: 2024-11-19フリードマン-ミンツボリス
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
JP2023571852
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Priority Date
2021-05-18
Filing Date
2022-05-13
Publication Date
2024-11-19
Estimated Expiration
2042-05-13

AI Technical Summary

Technical Problem

Existing speech processing technologies struggle to accurately recognize and synthesize human utterances, particularly in identifying and generating harmonic components that convey phonetic chords, due to interference and attenuation issues.

Method used

A computer-implemented method analyzes harmonic acoustic spectra from human speech to identify fundamental and harmonic components, using detection thresholds and primary cap frequencies to select phonetic chords, and synthesizes these chords by applying spectral data to an electroacoustic transducer.

Benefits of technology

Enhances the recognition and synthesis of human utterances by accurately identifying and generating harmonic phonetic chords, improving speech processing systems' accuracy and fidelity.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 00000000_0000_ABST
    Figure 00000000_0000_ABST
Patent Text Reader

Abstract

Within each harmonic spectrum of the set of spectra obtained from the analysis of waveforms representing human speech, two or more fundamental or harmonic components are identified whose frequencies are separated by integer multiples of the fundamental acoustic frequency. The highest harmonic frequency greater than 410 Hz is the primary cap frequency and is used to select a primary speech tone corresponding to a subset of speech chords from the set of speech chords for which the acoustic spectrum is available. The spectral data may also include frequencies of primary band, secondary band (or secondary tone), base band, or reduced base band acoustic components and may be used to select a speech chord from the subset of speech chords corresponding to the selected primary tone.
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The field of the invention relates to the recognition or synthesis of human vocalized speech. In particular, a computer-implemented method for recognizing or synthesizing human vocalized harmonic sounds is disclosed. [Background technology]

[0002] Some examples of speech processing or synthesis devices or methods are disclosed below. - U.S. Patent No. 5,406,635, issued April 11, 1995 to Jarvinen, entitled "Noise attenuation system." - U.S. Patent Application Publication No. 2004 / 0181411, issued on September 16, 2004 in the name of Gao and entitled "Voicing index controls for CELP speech encoding." - U.S. Patent No. 6,925,435, issued on August 2, 2005 to Gao, entitled “Method and apparatus for improved noise reduction in a speech encoder.” - U.S. Patent No. 7,516,067, issued on April 7, 2009 to Seltzer et al., entitled “Method and Apparatus Using Harmonic-Model-Based Front End for Robust Speech Recognition.” - U.S. Patent No. 7,634,401, issued to Fukuda on December 15, 2009, and entitled "Speech recognition method for determining missing speech." - U.S. Patent No. 8,566,088, issued on October 22, 2013 to Pinson et al., entitled “System and method for automatic speech to text conversion.” - U.S. Patent No. 8,606,566, issued on December 10, 2013 to Li et al., entitled “Speech enhancement through partial speech reconstruction.” - U.S. Patent No. 8,812,312, issued on August 19, 2014 to Fukuda et al., entitled “System, method and program for speech processing.” - U.S. Patent No. 9,087,513, issued on July 21, 2015 to Ichikawa et al., entitled “Noise reduction method, program product, and apparatus.” - U.S. Patent No. 9,190,072, issued on November 15, 2015 to Ichikawa and entitled “Local peak weighted-minimum mean square error (LPW- MMSE) estimation for robust speech.” - U.S. Patent No. 9,570,072, issued to Pinson on February 14, 2017, and entitled “System and method for noise reduction in processing speech signals by targeting speech and disregarding noise.” - Dieter Maurer, Acoustic of the Vowel: Preliminaries, Peter Lang AG, Bern 2016. - Bruno H. Repp, Categorical Perception: Issues, Methods, Findings, SPEECH AND LANGUAGE: Advances in Basic Research and Practice, Vol. 10 p. 243 (Academic Press 1984), https: / / doi.org / 10.1016 / B978-0-12-608610-2.50012-1. - U.S. Patent No. 9,147,393, issued on September 29, 2015 to Fridman-Mintz (the inventor of the present invention), entitled "Syllable based speech processing method." - U.S. Patent No. 9,460,707, issued on October 4, 2016 to Fridman-Mintz (the inventor of the present invention), entitled “Method and Apparatus for Electronically Recognizing a Series of Words Based on Syllable-Defining Beats.” - U.S. Patent No. 9,747,892, issued on August 29, 2017 to Fridman-Mintz (the inventor of the present invention), entitled “Method and Apparatus for Electronically Synthesizing Acoustic Waveforms Representing a Series of Words Based on Syllable-Defining Beats.”

[0003] The last three patents listed above (each to Fridman-Mintz) are incorporated by reference as if set forth in their entirety herein. Summary of the Invention

[0004] A computer-implemented method is employed to identify one or more audio chords (e.g., harmonic phones) represented in a time-lapse electronic waveform derived from human speech. In one embodiment, a time sequence of audio spectra derived from the waveform is analyzed to identify each of a plurality of harmonic audio spectra in the time sequence, each of which has an intensity above a detection threshold. The identified components have frequencies separated by at least one integer multiple of a fundamental audio frequency associated with the audio spectrum. For at least some of the plurality of audio spectra, a primary cap frequency is identified, the primary cap frequency being greater than 410 Hz and also the highest harmonic frequency among the identified harmonic components. For each audio spectrum for which a primary cap frequency is identified, the identified primary cap frequency is used to select at least one audio sound from a set of voices as a primary audio sound. The selected primary audio sounds correspond to a subset of audio chords from the set of audio chords.

[0005] In some implementations, the acoustic spectrum of the time sequence may correspond to one of a sequence of temporal sample intervals of the waveform, while in other implementations the acoustic spectrum corresponds to one of a sequence of distinct temporal segments in which the time-dependent acoustic spectrum of the waveform remains consistent with a single audio chord. In some cases, the audio chord may be selected based on harmonic content present in a harmonic audio spectrum that includes one or more of a primary band, a secondary band, a baseband, or a reduced baseband (each of which is described below).

[0006] A computer-implemented method is employed to analyze human vocalized speech and generate spectral data that can be used to identify harmonic vocal chords in the above method. For each vocal chord, a waveform obtained from vocalization of that vocal chord by one or more human subjects is spectrally analyzed. The spectral analysis includes, for each electronic waveform, an estimation of a fundamental acoustic frequency and the identification of two or more fundamental or harmonic components, each having an intensity above a detection threshold and an acoustic frequency that is the fundamental acoustic frequency or a harmonic thereof. A primary cap frequency is identified for each vocal chord and stored along with the acoustic frequency of each identified fundamental or harmonic component. A focal frequency can be estimated for a primary voice common to a subset of vocal chords using the observed primary cap frequencies (e.g., average, or median). In some cases, the stored spectral data can include data for one or more of a primary band, a secondary band, a base band, or a reduced base band.

[0007] A computer-implemented method is employed for synthesizing a temporal segment of an electronic waveform. An audio chord is generated by applying the waveform segment to an electro-acoustic transducer. A primary voice corresponding to a selected audio chord and data indicative of a focal frequency of the primary voice are used to determine a primary cap frequency. The primary cap frequency is (i) an integer multiple of the selected fundamental frequency, (ii) greater than 410 Hz, and (iii) closer to the focal frequency of the corresponding tonic voice than to the focal frequencies of other voices. A harmonic component of the primary cap frequency is included in the waveform segment. The primary cap frequency is the highest frequency among the harmonic components included in the waveform segment. The waveform segment may further include components at one or more harmonic frequencies of the primary band, secondary band, base band, or reduced base band. The method may be repeated for each audio chord in the temporal sequence constituting human speech, along with a plurality of different harmonic or hybrid segments, non-harmonic or unvoiced segments, and transitional segments therebetween.

[0008] Objects and advantages of the present invention relating to human speech recognition or synthesis will become apparent with reference to the exemplary embodiments illustrated in the drawings and disclosed in the following description or appended claims.

[0009] This Summary is provided to introduce some concepts in a simplified form that are described later in the Detailed Description. This Summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used as an aid in determining the scope of the claimed subject matter. [Brief description of the drawings]

[0010] [Figure 1] Figure 1 shows the recorded natural

number

number

number

number

number

number

number

number

[0011] The examples or embodiments depicted in the drawings are only shown in schematic form, not all features are shown in full detail or to scale, and certain features or structures may be exaggerated or reduced relative to other features or structures or may be omitted entirely for clarity. The drawings should not be considered to scale unless expressly indicated as such. The illustrated embodiments are illustrative only and should not be construed as limiting the scope of the disclosure or inventive subject matter. The same reference numbers refer to similar elements throughout the different figures.

[0012] The following detailed description should be read in conjunction with the drawings, which illustrate the principles of the subject matter by way of example and not by way of limitation.

[0013] The methods disclosed herein rely in part on the observation that for speech chords (e.g., harmonic tones such as vowels, nasal vowels, nasals, and approximants), harmonic acoustic spectral components (i.e., components with frequencies that are integer multiples of the fundamental acoustic frequency) are central to the linguistically relevant information that conveys those speech chords. Thus, the disclosed methods include recognition or synthesis of sonic chords using detection, identification, or generation of such harmonic components. Each speech corresponds to a fundamental chord element (e.g., as in the example list of FIG. 2), and each speech chord can include multiple chord frequencies in a given speech context (e.g., as in the example list of FIG. 3A and FIG. 3B).

[0014] As part of operating the speech recognition system, a computer-implemented method is employed for identifying one or more audio chords represented in an electronic waveform over time derived from a human voice utterance. A time sequence of audio spectra is derived from the waveform in a suitable manner (described below), and some of the audio spectra are identified as harmonic audio spectra, i.e., spectra that include frequency components at one or more integer multiples of a fundamental audio frequency. For some or all of the harmonic audio spectra in the time sequence, two or more fundamental or harmonic components are identified that have respective intensities above a detection threshold and are separated by integer multiples of the corresponding fundamental audio frequency of the harmonic audio spectrum. In some instances, a fundamental frequency component may be lost or hidden for various reasons, such as interference with other nearby higher intensity frequency components, attenuation due to obstructions in the sound path, poor production of low frequencies by small speakers, etc. Even if a fundamental component is not present in a harmonic spectrum, the harmonic spectrum may still be characterized by a corresponding fundamental audio frequency, the integer multiples of which separate the harmonic components of the harmonic spectrum.

[0015] Fundamental and harmonic components typically appear as peak-like spectral features above background levels (e.g., as in FIG. 1). The corresponding frequencies of the components may be defined in any suitable manner, e.g., as the frequency of maximum intensity or the centroid of the component. If the width of the spectral features is non-zero, some uncertainty may arise in assigning the frequencies corresponding to those features, and there is usually some leeway in determining whether a spectral feature is an integer multiple of the fundamental frequency. The intensity of the components may be characterized in any suitable manner, e.g., as maximum intensity, peak intensity, or integrated intensity. Intensity is usually expressed in relative terms (e.g., decibels) relative to background noise level, hearing threshold level, or another reference level. Regardless of how the relative intensity is defined, an appropriate detection threshold may be selected to determine whether a given spectral feature has been "detected" and "identified" as a fundamental or harmonic component of a given harmonic spectrum. In some examples, a decibel level may be specified as being above a selected reference level or above the background noise level. In some instances (eg, as in FIG. 1), the detection threshold is frequency dependent and may therefore be different for fundamental and harmonic components of the spectrum.

[0016] For at least some of the harmonic sound spectra, a primary cap frequency is identified that is the highest harmonic frequency among the identified harmonic components, provided that the highest harmonic frequency is also greater than 410 Hz. For each harmonic sound spectrum for which such a primary cap frequency is identified, the primary cap frequency is used to select at least one speech sound from a set of fundamental speech sounds (e.g., as in the table of FIG. 2). Such a selection is possible because, by recording and analyzing numerous speech chords produced by humans, it has been observed that the primary cap frequency most closely correlates with the particular speech chord produced. The data collected to create the table of fundamental speech sounds (e.g., as in FIG. 2) may include, for each speech sound, a characteristic primary cap frequency (i.e., a focal frequency) and the primary symbol label with which it correlates. Other speech sound datasets may also be employed. The generation of such speech sound datasets is described below. Other speech recognition systems often rely on the fundamental frequency and possibly two or three formants (time-aligned intensity peaks with frequency values ​​independent of the fundamental frequency) while substantially ignoring any harmonic components present in the acoustic spectrum, under the assumption that vowel perception and recognition is not based on categorical perception or feature analysis. The method disclosed herein utilizes these previously unexploited harmonic spectral components to identify a corresponding speech chord and one or more of its constituent voices, demonstrating that vowel identification can potentially be based on prototypical categorical perception.

[0017] For a given voice chord uttered at different pitches (i.e., different fundamental frequencies), the primary cap frequency may vary, so definitive identification (e.g., unambiguous selection from the data table of FIG. 2) cannot always be achieved. The tables of FIGS. 3A and 3B show examples of instantiated chord data. Each instantiated primary cap frequency falls between the focal frequencies of a pair of adjacent voice sounds (e.g., Table 2) or is otherwise higher than the highest focal frequency. If it falls between a pair of voice sounds (labeled the "identified" chord and the "adjacent" chord in the tables of FIGS. 3A and 3B), the instantiated voice chord can be identified as corresponding to one of these two voice sounds (the "identified" chord in this example). If it exceeds the highest focal frequency (e.g., above 4160 Hz in the table of FIG. 2), the instantiated chord can be identified as corresponding to one of these two voice sounds (the "identified" chord in this example). TIFF2024518630000010.tif9170 may be identified as corresponding to a speech sound. In some examples, the instantiated speech chord is identified as corresponding to the speech sound of the pair of adjacent speech sounds having a focal frequency closer to the instantiated primary cap frequency. The confidence column in the tables of Figures 3A and 3B indicates the relative closeness of the instantiated primary cap frequency to the corresponding pair of adjacent focal frequencies. Other suitable tests or algorithms may be employed to determine different possible speech chords that may match a particular instantiated harmonic spectrum. In some examples, an artificial intelligence (Al) system or neural network may be trained to make the selection. In some examples, a so-called "fuzzy logic" algorithm may be employed. In some examples, identification of additional fundamental or harmonic spectral components of the observed harmonic spectrum (e.g., primary band, secondary band, base band, as described further below) may be employed to distinguish between different voices having focal frequencies that match the observed primary cap frequency, secondary cap frequency, or base cap frequency (described further below).

[0018] Some exemplary methods may include deriving a time sequence of acoustic spectra from a time-lapse electronic waveform, while in other examples the time sequence of acoustic spectra has already been derived from the electronic waveform prior to performing the method. In either case, the time sequence of acoustic spectra may be derived from the waveform in any suitable manner, such as by processing the electronic waveform itself using an electronic spectrum analyzer, or by processing a numerical representation of the waveform using Fourier transform techniques.

[0019] In some examples, each of the acoustic spectra corresponds to one of a sequence of temporal sample intervals of the waveform, where such sample intervals may, but need not, be of equal duration. The acoustic spectra of at least some of such temporal sequences may be classified as only one of harmonic, inharmonic, hybrid, or unvoiced. For at least some of the temporal sample intervals classified as harmonic, a method described above or below may be employed to identify an audio chord from among a set of audio chords (each of which includes a characteristic combination of notes) that corresponds to that temporal sample interval of the electronic waveform.

[0020] In some examples, each of the acoustic spectra corresponds to one of a set of different time segments whose time-dependent acoustic spectrum remains consistent with a single audio chord. A determination that an audio spectrum "remains consistent with a single audio chord" can be made without having previously identified a chord, simply by observing that no transitions to audio spectra indicative of different chords occur. At least some of the time segments can be classified as only one of harmonic, non-harmonic, hybrid, or unvoiced based on their acoustic spectra. For at least some of the time segments classified as harmonic, a method described above or below can be employed to identify an audio chord from a set of audio chords as corresponding to that time segment of the electronic waveform.

[0021] Identifying additional fundamental or harmonic frequencies in addition to the primary cap frequency can further facilitate identification of audio chords corresponding to the harmonic spectrum in the time sequence. In some cases, identifying additional fundamental or harmonic components can help distinguish audio chords that have similar primary cap frequencies. In some instances, the additional fundamental or harmonic components form one or more of the primary band, secondary band, or fundamental band.

[0022] In some examples, a primary band of harmonic components can be identified in at least some of the harmonic spectra of the time sequence. The primary band can include harmonic components at the primary cap frequency and one, two, three, or more highest consecutive multiples of the fundamental frequency that are (i) less than the primary cap frequency, (ii) greater than 410 Hz, and (iii) greater than the smallest integer multiple of the fundamental frequency greater than 410 Hz. The stored data for the audio chord set can include the frequencies of other harmonic components in the primary band in addition to the primary cap frequency. Based on a comparison of the data for the primary band with the observed frequencies of a particular harmonic spectrum obtained from the electronic waveform, an audio chord of the set can be selected as corresponding to the corresponding time portion of the harmonic spectrum and waveform. Figures 4A / 4B, 5, 6A / 6B, 7, and 8 show some examples of recorded and synthesized waveforms and harmonic spectra that can potentially be distinguished based on the primary band of harmonic components.

[0023] In some examples, a secondary band of harmonic components can be identified in at least some of the harmonic spectra of the time sequence. The secondary band can include harmonic components at one or more harmonic acoustic frequencies that are greater than the smallest integer multiple of the fundamental acoustic frequency above 410 Hz and that are separated from the primary cap frequency by at least one intervening multiple of the fundamental acoustic frequency where the acoustic spectrum lacks harmonic components. In other words, the secondary band falls below a "harmonic gap" or one or more "missing harmonics" between the lowest frequency component of the primary band and the highest frequency component of the secondary band. The frequency of the highest frequency component of the secondary band may be referred to as the secondary cap frequency.

[0024] The data stored for a set of audio chords may include, in addition to the primary cap frequency (or corresponding primary symbol), the secondary cap frequency (or corresponding secondary symbol) and the frequency (if any) of one or more other harmonic components of the secondary band. Based on a comparison of that secondary band data to the observed frequency of a particular harmonic spectrum obtained from an electronic waveform, an audio chord of the set may be selected as corresponding to that harmonic spectrum and the corresponding time portion of the waveform. For example, by observing the secondary band, it may be possible to distinguish between (i) a first audio chord having a secondary band separated from the primary band by only one missing harmonic and (ii) a second audio chord having a secondary band separated from the primary band by two or more missing harmonics. In some examples, the comparison of the secondary band data to the observed components may be used in conjunction with the comparison of the primary band data to the observed components. In other examples, the secondary band data and components may be used without the use of the primary band data and components. Figures 4A / 4B, 5, 6A / 6B, 7, and 8 show some examples of recorded and synthesized waveforms and harmonic spectra that may be distinguished based on the secondary bands of the harmonic components.

[0025] In some examples, a base band of harmonic components can be identified in at least some of the harmonic spectra of the time sequence. The base band can include harmonic components at one or more fundamental or harmonic acoustic frequencies below 410 Hz, and can also include a harmonic component at a lowest harmonic acoustic frequency above 410 Hz (unless that harmonic frequency is the primary cap frequency). The frequency of the highest frequency component of the base band can be referred to as the base cap frequency. In examples where the primary cap frequency is also the only harmonic frequency above 410 Hz, the harmonic spectrum includes only the primary cap frequency and the base band components, and no other primary band components or secondary band components. The stored chord set data can include, in addition to the primary cap frequency (or corresponding primary audio tone), the fundamental cap frequency and the frequencies of one or more other harmonic components of the fundamental band. Based on a comparison of the fundamental band data with the observed frequencies of a particular harmonic spectrum obtained from the electronic waveform, an audio chord of the set can be selected as corresponding to the corresponding time portion of the harmonic spectrum and waveform.

[0026] In some examples, a comparison of the baseband data to the observed components may be used in combination with a comparison of the primary band data to the observed components, in other examples, the baseband data and components may be used in combination with a comparison of the secondary band data to the observed components, in other examples, the baseband data and components may be used in combination with a comparison of the observed components to both the primary band data and the secondary band data, and in other examples, the baseband data and components may be used without either the primary band data or the secondary band data. Figures 4A / 4B, 5, 6A / 6B, 7, and 8 show some examples of recorded and synthesized waveforms and harmonic spectra that may be distinguishable based on the baseband of the harmonic components.

[0027] In some examples, the harmonic acoustic spectrum of a temporal sequence may include only baseband components with upper frequency limits below 410 Hz. Such harmonic acoustic spectra are referred to as reduced baseband and may correspond to particular harmonic acoustic schemata (e.g., nasals, nasalized vowels, or approximants) or hybrid acoustic schemata (e.g., voiced fricatives). The stored data for these acoustic schemata may include frequencies of the reduced baseband harmonic components. Based on a comparison of the reduced baseband data with the observed frequencies of particular harmonic spectra obtained from the electronic waveform, a set of harmonic or hybrid acoustic schemata may be selected as corresponding to the harmonic spectrum and corresponding temporal portion of the waveform. The presence or absence of high frequency non-harmonic frequency components may also be determined by (i) reduced baseband harmonic schemata (e.g., corresponding to the first column of FIG. 2) and (ii) hybrid schemata (e.g., This can be used to distinguish between voiced fricatives such as TIFF2024518630000011.tif13170.

[0028] Various methods disclosed herein for recognizing speech tones and chords in enunciated human speech rely on stored data indicating harmonic spectra for each of a number of speech tones or chords, including expected harmonic frequencies for each speech tone or chord (e.g., the tables of Figs. 2, 3A, and 3B). A method for generating that data for a set of speech tones and chords may include performing acoustic spectrum analysis on multiple utterances of the speech chord by one or more human subjects instructed to articulate certain sounds or words, while retaining the labeling information of the instructed pronunciations. Such a process is similar to "teaching" a system to recognize each speech chord as it occurs in a waveform representing human speech. In some examples, harmonic spectrum data may be generated based on the speech of a single human subject to train the system to recognize the speech of a particular subject. In other examples, speech spectrum data may be generated based on an average or other analysis of the speech of multiple human subjects to train the system for more general speech recognition.

[0029] For each voice tone or chord, the spectral analysis includes estimating a fundamental acoustic frequency and identifying two or more fundamental or harmonic components detected or identified in the spectrum. As described above, a component that is "detected" or "identified" is one that has an intensity above one or more appropriately defined detection thresholds. Each component has an acoustic frequency that is a fundamental acoustic frequency or a harmonic acoustic frequency (i.e., an integer multiple of the fundamental acoustic frequency). Although a spectrum is characterized by a fundamental acoustic frequency, the spectrum may or may not include a fundamental component at that fundamental acoustic frequency. Of the identified harmonic components, the highest harmonic acoustic frequency greater than 410 Hz is identified as the primary cap frequency, and that frequency is stored as part of the data for the voice code. Also, as described above, the acoustic frequency of each identified fundamental or harmonic component is stored. For some chords, the data may also include the acoustic frequencies of one or more identified harmonic components in the primary, secondary, and fundamental bands, as described above.

[0030] To estimate the focal frequency of a speech sound, a subset of speech chords is produced multiple times by one or more subjects at multiple different fundamental frequencies and spectrally analyzed. The speech chords in the subset share a common tonic. The focal frequency of the common tonic can be estimated from the primary cap frequencies of the produced speech chords. The focal frequency can be estimated from the observed primary cap frequencies in any suitable manner, such as the mean or median of the observed primary cap frequencies.

[0031] For some harmonic acoustic schemata (e.g., nasals, nasalized vowels, or approximants) or hybrid acoustic schemata (e.g., voiced fricatives), the spectral analysis may identify fundamental or harmonic components only at frequencies below 410 Hz, as noted above as reduced baseband frequencies. The acoustic frequencies of these reduced baseband components may be included in a data set describing the corresponding harmonic or hybrid acoustic schemata, along with an indicator of the absence or presence, respectively, of higher frequency inharmonic components.

[0032] Recognizing the importance of harmonic components in the recognition of speech tones and chords in human vocalized speech also allows for improved speech synthesis. The above-described spectral data used to recognize chords can also be used to generate chords. To synthesize a selected speech chord, a primary cap frequency can be determined based on a corresponding tonic (primary) note and a selected fundamental frequency (i.e., pitch). For a harmonic speech chord, the primary cap frequency (i) is an integer multiple of the selected fundamental frequency, (ii) is greater than 410 Hz, and (iii) is closer to the corresponding tonic focus frequency than the focus frequency of the other speech tones. The frequency components at the tonic frequency are included in the synthesized waveform segment. In some examples, if the selected speech chord includes secondary tones, the secondary cap frequency can be determined as described above for the tonic and cap frequencies, and the frequency components at the secondary cap frequency can be included in the synthesized waveform segment.

[0033] This method can be repeated for each set of chords and fundamental frequencies in a time sequence interspersed with a number of different harmonic segments (e.g., vowels, nasals, nasals, approximants) and hybrid segments (e.g., voiced fricatives). The synthesized harmonic segments, together with synthesized non-harmonic segments (e.g., voiceless fricatives), synthesized unvoiced segments (e.g., obstruents, trills, flaps), and transitional segments therebetween, constitute synthesized human speech. The electronic waveforms thus generated using the spectral data are applied to an electro-acoustic transducer (e.g., a speaker) to generate synthesized speech chords. A string of such chords can be generated to constitute a synthetic speech sequence.

[0034] In some examples, to generate a speech chord for which spectral data is available, an electronic waveform corresponding to the chord is created using the corresponding spectral data (primary cap frequency along with the acoustic frequencies of the harmonic components). For some speech sounds and chords, the data may include one or more identified acoustic frequencies of the primary, secondary, and baseband harmonic components, as described above. For some harmonic or hybrid acoustic schemata (e.g., speech chords lacking primary and secondary components, such as nasals and voiced fricatives), the data may include an indication of the presence or absence of reduced baseband components and high frequency non-harmonic components.

[0035] The systems and methods disclosed herein may be implemented as or with a general purpose or special purpose computer, server, or other programmable hardware device programmed through software, or as hardware or equipment "programmed" through hardwiring, or a combination of the two. A "computer" or "server" may consist of a single machine or may consist of multiple interacting machines (at a single location or multiple remote locations). The computer program or other software code, if used, may be implemented in tangible, non-transitory, temporary or permanent storage or exchangeable media, such as including microcode, machine code, network-based or web-based or distributed software modules operating together, programming in RAM, ROM, CD-ROM, CD-R, CD-R / W, DVD-ROM, DVD±R, DVD±R / W, hard drive, thumb drive, flash memory, optical media, magnetic media, semiconductor media, or any future computer readable storage alternative. An electronic indicia of a data set may be read from, received from, or stored in any of the tangible and non-transitory computer readable media mentioned herein.

[0036] In addition to the above, the following exemplary embodiments are within the scope of the present disclosure or the appended claims.

[0037] Example 1. A computer-implemented method for identifying one or more vocal chords represented in a time-lapse electronic waveform derived from a human vocal utterance, the method comprising using one or more electronic processors of a computer system programmed to: (a) for each of a plurality of harmonic audio spectra of a time sequence of audio spectra derived from the waveform, identify two or more fundamental or harmonic components within the harmonic audio spectrum, each identified component having an intensity above a detection threshold, the identified components having frequencies separated by at least one integer multiple of a fundamental audio frequency associated with the audio spectrum, (b) for at least some of the plurality of audio spectra, identify a highest harmonic frequency among the identified harmonic components as a primary cap frequency, the highest harmonic frequency also being greater than 410 Hz, and c) for each of the plurality of audio spectra for which a primary cap frequency is identified in part (b), select at least one voice from a set of voices as a primary voice using the identified primary cap frequency, the selected primary voices corresponding to a subset of vocal chords in the set of vocal sounds.

[0038] Example 2. The method of Example 1, wherein each of the acoustic spectra corresponds to one of a sequence of temporal sample intervals of the waveform, further comprising using one or more of the electronic processors of the programmed computer system to: (A) classify at least some of the acoustic spectra of the temporal sequences as only one of harmonic, inharmonic, hybrid, or unvoiced; and (B) perform parts (a)-(c) for at least some of the temporal sample intervals classified as harmonic in part (A) to identify a selected tonic note as corresponding to that temporal sample interval of the electronic waveform.

[0039] Example 3. The method of Example 1, wherein each of the acoustic spectra corresponds to one of a series of distinct time segments in which the time-dependent acoustic spectrum of the waveform remains consistent with a single audio chord, further comprising using one or more electronic processors of a programmed computer system to: (A) classify at least some of the time segments as only one of harmonic, inharmonic, or hybrid based on their acoustic spectra; and (B) perform parts (a)-(c) for at least some of the time segments classified as harmonic in part (A) to identify a selected tonic note as corresponding to that time segment of the electronic waveform.

[0040] Example 4. The method of any of Examples 1 to 3, further comprising deriving a time sequence of acoustic spectra from the time-lapse electronic waveform using one or more electronic processors of a programmed computer system.

[0041] Example 5. The method of any one of Examples 1 to 4, wherein the detection thresholds for one or more different fundamental or harmonic components vary depending on the acoustic frequency.

[0042] Example 6. The method of any of Examples 1 to 5, further comprising using one or more of the electronic processors of the programmed computer system to identify, for at least one of the harmonic sound spectra of the temporal sequence of sound spectra derived from the waveform, a fundamental component within the harmonic sound spectrum having a fundamental sound frequency and an intensity above a detection threshold.

[0043] Example 7. The method of any of Examples 1-6, further comprising using one or more of the electronic processors of the programmed computer system to select a sonic chord from among the subset of sonic chords of part (c) for a particular harmonic sound spectrum of the plurality of sound spectra based at least in part on a comparison of (i) the stored data indicating expected harmonic frequencies for the subset of sonic chords and (ii) the harmonic frequencies of harmonic components of a primary band of the harmonic sound spectrum, the primary band including a primary cap frequency and harmonic components at 1, 2, or 3 greatest consecutive multiples of the fundamental sound frequency that are less than the primary cap frequency and greater than 410 Hz, and greater than a minimum integer multiple of the fundamental sound frequency that is greater than 410 Hz and less than the primary cap frequency.

[0044] Example 8. The method of any of Examples 1-7, further comprising using one or more of the electronic processors of the programmed computer system to select, for a particular harmonic sound spectrum of the plurality of sound spectra, a sonic chord from among the subset of part (c) based at least in part on a comparison of (i) the stored data, the data indicating expected harmonic frequencies for the sonic chords in the set, and (ii) the harmonic frequencies of the harmonic components of a sub-band of the harmonic sound spectrum, the sub-band containing harmonic components at one or more harmonic sound frequencies greater than a smallest integer multiple of the fundamental sound frequency that is greater than 410 Hz and separated from the primary cap frequency by at least one intervening multiple of the fundamental sound frequency in which the sound spectrum is devoid of harmonic components.

[0045] Example 9. The method of any of Examples 1-8, further comprising using one or more of the electronic processors of the programmed computer system to, for a particular harmonic sound spectrum of the plurality of sound spectra, base the selection of part (c) at least in part on a comparison of (i) the stored data indicating expected harmonic frequencies for the set of vocal chords and (ii) harmonic frequencies of harmonic components of a base band of the harmonic sound spectrum, the base band including harmonic components at one or more fundamental or harmonic sound frequencies that are less than 410 Hz or equal to the smallest integer multiple of the fundamental sound frequency that is between 410 Hz and a primary cap frequency.

[0046] Example 10. The method of any of Examples 1-9, further comprising, for a selected harmonic acoustic spectrum in which the highest harmonic acoustic frequency of an identified harmonic component is less than 410 Hz, using one or more electronic processors of a programmed computer system to select one of the schemes from the set of harmonic or hybrid acoustic schemes based at least in part on (i) comparing (A) the compiled data indicating expected harmonic frequencies for the set of acoustic schemes with (B) the harmonic frequency of each identified fundamental or harmonic component, and (ii) the presence or absence of higher frequency non-harmonic frequency components.

[0047] Example 11. A computer-implemented method for generating stored data indicative of expected harmonic frequencies for a harmonic acoustic spectrum of each of a set of audio chords, the method comprising using one or more electronic processors of a computer system programmed to: (a) for each audio chord of the set, spectrally analyze a plurality of electronic waveforms obtained when one or more human subjects utter that audio chord, the spectral analysis including, for each electronic waveform, an estimation of a fundamental audio frequency and identification of two or more fundamental or harmonic components, each having an intensity above a detection threshold and an audio frequency that is the fundamental audio frequency or a harmonic audio frequency equal to an integer multiple of the fundamental audio frequency; and (b) identify, among the identified harmonic components, the highest harmonic audio frequency greater than 410 Hz as a primary cap frequency, wherein, for each audio chord for which a primary cap frequency is thus identified, storing, in a tangible, non-transient computer-readable storage medium that is not a transient propagating signal, an electronic indication of the primary cap frequency and the audio frequency of each identified fundamental or harmonic component.

[0048] Example 12. The method of Example 11, further comprising: for a subset of vocal chords having a common primary voice, and for a plurality of utterances at a plurality of different fundamental frequencies by one or more human subjects, (i) estimating a focal frequency of the common primary voice corresponding to the subset of vocal chords from the primary cap frequency, and (ii) storing an electronic indication of the focal frequency of the primary voice in a tangible, non-transient, computer-readable storage medium that is not a transient, propagated signal.

[0049] Example 13. The method of any of Examples 11 or 12, wherein the detection threshold varies as a function of acoustic frequency.

[0050] Example 14. The method of any of Examples 11 to 13, wherein the stored data includes, for at least some of the audio chords of the set, harmonic frequencies of harmonic components of a primary band of a corresponding harmonic audio spectrum, the primary band including a primary cap frequency and one, two, three or more largest consecutive multiples of the fundamental audio frequency that are (i) less than the primary cap frequency, (ii) greater than 410 Hz, and (iii) greater than the smallest integer multiple of the fundamental audio frequency that is greater than or equal to 410 Hz and less than the primary cap frequency.

[0051] Example 15. The method of any of Examples 11 to 14, wherein the stored data includes, for at least some of the audio chords of the set, harmonic frequencies of harmonic components of a secondary band of a corresponding harmonic audio spectrum, the secondary band including one or more harmonic components at a harmonic audio frequency greater than a smallest integer multiple of the fundamental audio frequency greater than 410 Hz, the audio spectrum being separated from the primary cap frequency by at least one intervening multiple of the fundamental audio frequency that is devoid of harmonic components.

[0052] Example 16. The method of any of Examples 11 to 15, wherein the stored data includes, for at least some of the audio chords of the set, harmonic frequencies of harmonic components of a baseband of a corresponding harmonic audio spectrum, the baseband including one or more harmonic components at one or more fundamental or harmonic audio frequencies that are less than 410 Hz or equal to the smallest integer multiple of the fundamental audio frequency that is between 410 Hz and a primary cap frequency.

[0053] Example 17. The method of any of Examples 11-16, further comprising: (a) for each one of the additional harmonic or hybrid acoustic schemata, spectrally analyzing a plurality of electronic waveforms obtained when the one or more human subjects speak that schema, where for each electronic waveform, the spectral analysis includes estimating a fundamental acoustic frequency and identifying two or more fundamental or harmonic components, each having an intensity above a detection threshold and having a fundamental acoustic frequency or a harmonic acoustic frequency equal to an integer multiple of the fundamental acoustic frequency, where each of the fundamental and harmonic acoustic frequencies is less than 410 Hz; and b) for each acoustic schema for which the plurality of waveforms are analyzed in part (a), storing in a tangible and non-transient computer-readable storage medium that is not a transient propagating signal an electronic indication of the presence or absence of one or more fundamental and harmonic acoustic frequencies of each identified fundamental or harmonic component, and a higher frequency non-harmonic component.

[0054] Example 18. A computer-implemented method for synthesizing a temporal segment of an electronic waveform and producing the sound of a selected audio chord from among a set of audio chords when the waveform segment is applied to an electro-acoustic transducer, the method comprising: (a) using received, acquired, or calculated data indicative of a tonic note corresponding to a selected audio chord and a focal frequency of the tonic note, determining a primary cap frequency using the focal frequency and a selected fundamental audio frequency, the primary cap frequency being (i) an integer multiple of the selected fundamental frequency, (ii) greater than 410 Hz, and (iii) closer to the corresponding tonic focus frequency than the focus frequencies of other audio chords; and b) including in the waveform segment a harmonic component of the primary cap frequency, the primary cap frequency being greater than audio frequencies of all other harmonic components included in the waveform segment.

[0055] Example 19. The method of Example 18, further comprising repeating the method of Example 18 for each voice chord in a temporal sequence of a plurality of different harmonic or hybrid segments, along with inharmonic or unvoiced segments and transitional segments therebetween, that together comprise a human utterance.

[0056] Example 20. The method of any of Examples 18 or 19, further comprising applying waveform segments to an electro-acoustic transducer to produce sounds of the selected audio chord.

[0057] Example 21. The method of any of Examples 18-20, further comprising, for one or more audio chords of the set, including in the waveform segment a primary band of harmonic components, the primary band including harmonic components at the primary cap frequency and one, two, three, or more greatest consecutive multiples of the fundamental acoustic frequency that are (i) less than the primary cap frequency, (ii) greater than 410 Hz, and (iii) greater than the smallest integer multiple of the fundamental acoustic frequency that is greater than or equal to 410 Hz and less than the primary cap frequency.

[0058] Example 22. The method of any of Examples 18-20, further comprising, for one or more audio chords of the set, including in the waveform segment a primary band of harmonic components, the primary band including harmonic components of a primary cap frequency and at least three highest consecutive multiples of the fundamental audio frequency that are (i) less than the primary cap frequency, (ii) greater than 410 Hz, and (iii) greater than the smallest integer multiple of the fundamental audio frequency that is greater than 410 Hz and less than the primary cap frequency.

[0059] Example 23. The method of any of Examples 18-22, further comprising: for one or more audio chords of the set, (a) using the received, retrieved, or calculated data indicating a secondary audio wave corresponding to the selected audio chord and a focal frequency of the secondary audio wave, determining a secondary cap frequency using the focal frequency and the selected fundamental audio frequency, the secondary cap frequency being (i) an integer multiple of the selected fundamental audio frequency, (ii) greater than 410 Hz, and (iii) closer to the focal frequency of the corresponding secondary audio chord than the focal frequency of any other audio chord; and b) including in the waveform segment one or more harmonic components of a secondary band, the secondary band including harmonic components of a harmonic audio frequency that is (i) greater than a smallest integer multiple of the fundamental audio frequency that is greater than 410 Hz, and (ii) less than or equal to the secondary cap frequency.

[0060] Example 24. The method of any one of Examples 18 to 23, further comprising, for one or more audio chords of the set, including in the waveform segment one or more harmonic components of a baseband, the baseband including harmonic components at one or more fundamental or harmonic acoustic frequencies (i) less than 410 Hz, or (ii) equal to the smallest integer multiple of the fundamental acoustic frequency that is between 410 Hz and a primary cap frequency.

[0061] Example 25. The method of any of Examples 18-24, further comprising, for at least one additional time segment of the electronic waveform corresponding to a selected harmonic or hybrid acoustic schema of the set of such acoustic schema, including in the additional waveform segment one or more harmonic components of a reduced baseband only, the reduced baseband including only harmonic components at one or more fundamental or harmonic acoustic frequencies that are less than 410 Hz.

[0062] Example 26. The method of any of Examples 18-25, further comprising, for at least one additional time segment of the electronic waveform corresponding to a selected harmonic or hybrid acoustic schema of the set of such acoustic schema, including in the additional waveform segment (i) one or more reduced baseband only harmonic components, the reduced baseband including only harmonic components at one or more fundamental or harmonic acoustic frequencies below 410 Hz, and (ii) one or more higher frequency non-harmonic components.

[0063] Example 27. A programmed computerized machine including one or more electronic processors and one or more tangible computer-readable storage media operatively coupled to the one or more processors, the machine configured and programmed to perform the method of any of Examples 1 to 26.

[0064] Example 28. An article including a tangible medium that is not a transitory propagating signal, the medium being encoded with computer-readable instructions that, when applied to a computer system, instruct the computer system to perform any of the methods of Examples 1 to 26.

[0065] The present disclosure is illustrative and not restrictive. Further modifications will be apparent to those skilled in the art in light of the present disclosure and are intended to fall within the scope of the present disclosure. Equivalents of the disclosed exemplary embodiments and methods, or modifications thereof, are intended to be included within the scope of the present disclosure.

[0066] In the foregoing detailed description, various features may be grouped together in some exemplary embodiments for the purpose of streamlining the disclosure. This method of disclosure is not to be construed as reflecting an intention that a particular embodiment requires more features than are expressly recited therein. Rather, inventive subject matter may lie in less than all features of a single disclosed exemplary embodiment. Thus, the present disclosure is to be construed as implicitly disclosing any embodiment having any preferred subset of one or more features shown, described, or specified in the present application. A "preferred" subset of features includes only those features that are not incompatible or mutually exclusive with respect to other features of that subset. Furthermore, it should be noted that the cumulative scope of the examples listed above may, but does not necessarily, encompass the entirety of the subject matter disclosed in the present application.

[0067] For purposes of this disclosure, the following interpretations shall apply: "comprising," "including," "having," and variations thereof, wherever they occur, shall be construed as open-ended terms having the same meaning as if each instance of the word "at least" were added, unless expressly stated otherwise. Singular nouns shall be construed as one or more, unless "only," "single," or other similar limitation is expressly stated or implied by the particular context. The conjunction "or" shall be construed inclusively unless (i) it is expressly stated otherwise, for example, by use of "either," "only one of," or similar wording, or (ii) two or more of the listed alternatives are understood or disclosed (implied or explicit) as being incompatible or mutually exclusive in the particular context. In the latter case, "or" shall be understood to encompass only combinations that include alternatives that are not mutually exclusive. In one example, each of "a dog or a cat," "one or more of a dog or a cat," and "one or more dogs or cats" is to be interpreted as one or more dogs (no cats), one or more cats (no dogs), or one or more of each. As another example, each of "a dog, a cat, or a rat", "one or more of a dog, a cat, or a rat", and "one or more dogs, cats, or rats" is to be interpreted as (i) one or more dogs (no cats or rats), (ii) one or more cats (no dogs or rats), (iii) one or more rats (no dogs or cats), (iv) one or more dogs and one or more cats (no rats), (v) one or more dogs and one or more rats (no cats), (vi) one or more cats and one or more rats (no dogs), and (vii) one or more dogs, one or more cats, and one or more rats.As another example, "two or more of a dog, a cat, or a rat" or "two or more dogs, cats, or rats" is to be interpreted as (i) one or more dogs and one or more cats (no rats), (ii) one or more dogs and one or more rats (no cats), (iii) one or more cats and one or more rats (no dogs), or (iv) one or more dogs and one or more cats and one or more rats. "Three or more," "four or more," etc. are to be interpreted similarly.

[0068] When terms such as "approximately equal," "substantially equal," "approximately greater," "approximately less," and the like are used in connection with numerical quantities in this disclosure or in the appended claims, standard practices regarding measurement precision and significant figures shall apply unless a different interpretation is expressly provided. For ineffective quantities described with expressions such as "substantially prevented," "substantially absent," "substantially eliminated," "approximately equal to zero," "negligible," and the like, each such expression shall indicate a case in which the quantity in question has been reduced or diminished to such an extent that, for practical purposes in the context of the intended operation or use of the disclosed or claimed device or method, the overall behavior or performance of the device or method will be no different than would have occurred if the ineffective quantity had in fact been completely eliminated, was precisely equal to zero, or was otherwise precisely neutralized.

[0069] For purposes of this disclosure, any labeling of elements, steps, limitations, or other parts of an embodiment or example (e.g., first, second, third, etc., (a), (b), (c), etc., or (i), (ii), (iii), etc.) is used for purposes of clarity only and is not to be construed as implying any ordering or priority of the parts so labeled. If such ordering or priority is intended, it will be explicitly stated in the embodiment or example, or, in some cases, will be implicit or inherent based on the specific content of the embodiment or example. If it is desired to invoke the provisions of 35 U.S.C. 112(f) or corresponding statutes relating to "means+function" or "step+function" claim formats in an apparatus description, the word "means" will be stated in the description. If it is desired to invoke these provisions in a method description, the description will include the word "step". Conversely, if the words "means" or "step" do not appear, invocation of such provisions is not intended.

[0070] If any one or more disclosures are incorporated herein by reference and such incorporated disclosures conflict in whole or in part with this disclosure or differ in scope, then this disclosure shall control to the extent of the conflict, broader disclosure, or broader definition of a term. If any or all of the incorporated disclosures conflict with this disclosure, then the later-disclosed disclosure shall govern to the extent of the conflict.

[0071] The Abstract is provided as an aid in searching for particular subject matter within the patent document, however, the Abstract is not intended to imply that elements, features, or limitations described therein are necessarily included in, or in any way required by, any particular specification.

Claims

1. 1. A computer-implemented method for identifying one or more audio chords represented in a time-lapse electronic waveform derived from human speech, comprising: (a) for each of a plurality of harmonic sound spectra of a time sequence of sound spectra derived from a waveform, identifying two or more fundamental or harmonic components within the harmonic sound spectrum, each identified component having an intensity above a detection threshold, the identified components having frequencies separated by at least one integer multiple of a fundamental sound frequency associated with the sound spectrum; (b) for at least some of the plurality of acoustic spectra, identifying a highest harmonic frequency among the identified harmonic components as a primary cap frequency, the highest harmonic frequency also being greater than 410 Hz; and (c) for each of a plurality of acoustic spectra for which a primary cap frequency has been identified in (b), selecting at least one speech sound from the set of speech sounds as a primary speech sound using the identified primary cap frequency, the selected primary speech sounds corresponding to a subset of speech chords in the set of speech sounds; using one or more electronic processors of a computer system programmed for:

2. each of the acoustic spectra corresponds to one of a sequence of temporal sample intervals of the waveform; (A) classifying the acoustic spectra of at least some of the temporal sequences as only one of harmonic, inharmonic, hybrid, or voiceless; (B) for at least some of the time sample intervals classified as harmonics in part (A), performing parts (a) through (c) to identify a selected tonic note as corresponding to that time sample interval of the electronic waveform; The method of claim 1 , further comprising using one or more of the electronic processors of a computer system programmed for:

3. each of the acoustic spectra corresponds to one of a series of distinct time segments during which the time-dependent acoustic spectrum of the waveform remains consistent with a single audio chord; (A) classifying at least some of the time segments as only one of harmonic, inharmonic, or hybrid based on their acoustic spectrum; and (B) for at least some of the time segments classified as harmonic in part (A), performing parts (a) through (c) to identify a selected tonic note as corresponding to that time segment of the electronic waveform; The method of claim 1 , further comprising using one or more electronic processors of a computer system programmed for:

4. 10. The method of claim 1, further comprising deriving a time sequence of acoustic spectra from the time-lapse electronic waveform using one or more electronic processors of a programmed computer system.

5. The method of claim 1 , wherein the detection thresholds for one or more different fundamental or harmonic components vary depending on the acoustic frequency.

6. 2. The method of claim 1, further comprising using one or more of the electronic processors of the programmed computer system to identify, for at least one harmonic sound spectrum of a temporal sequence of sound spectra obtained from the waveform, a fundamental component within the harmonic sound spectrum having a fundamental sound frequency and an intensity above a detection threshold.

7. 2. The method of claim 1, further comprising: using one or more of the electronic processors of the programmed computer system to select a sonic chord from among the subset of sonic chords of part (c) for a particular harmonic sound spectrum of the plurality of sound spectra based at least in part on a comparison of (i) the stored data indicating expected harmonic frequencies for the subset of sonic chords and (ii) the harmonic frequencies of harmonic components of a primary band of the harmonic sound spectrum, the primary band including the primary cap frequency and harmonic components at 1, 2, or 3 greatest consecutive multiples of the fundamental sound frequency that are less than the primary cap frequency and greater than 410 Hz, and greater than a minimum integer multiple of the fundamental sound frequency that is greater than 410 Hz and less than the primary cap frequency.

8. 2. The method of claim 1, further comprising: using one or more of the electronic processors of the programmed computer system to select, for a particular harmonic sound spectrum of the plurality of sound spectra, an audio chord from among the subset of part (c) based at least in part on a comparison of (i) the stored data, the data being indicative of expected harmonic frequencies for the audio chords in the set, and (ii) the harmonic frequencies of the harmonic components of a sub-band of the harmonic sound spectrum, the sub-band comprising harmonic components at one or more harmonic sound frequencies greater than a smallest integer multiple of the fundamental sound frequency greater than 410 Hz and separated from the primary cap frequency by at least one intervening multiple of the fundamental sound frequency in which the audio spectrum is devoid of harmonic components.

9. 2. The method of claim 1, further comprising using one or more of the electronic processors of the programmed computer system to make, for a particular harmonic sound spectrum of the plurality of sound spectra, the selection of part (c) based at least in part on a comparison of (i) the stored data indicating expected harmonic frequencies for the set of vocal chords and (ii) harmonic frequencies of harmonic components of a base band of the harmonic sound spectrum, the base band including harmonic components at one or more fundamental or harmonic sound frequencies that are less than 410 Hz or equal to the smallest integer multiple of the fundamental sound frequency that is between 410 Hz and a primary cap frequency.

10. 10. The method of claim 1, further comprising: for a selected harmonic acoustic spectrum in which the highest harmonic acoustic frequency of an identified harmonic component is less than 410 Hz, using one or more electronic processors of a programmed computer system to select one of the harmonic or hybrid acoustic schemata from the set of those schemata based at least in part on (i) comparing (A) the stored data indicative of expected harmonic frequencies for the set of acoustic schemata with (B) the harmonic frequency of each identified fundamental or harmonic component, and (ii) the presence or absence of higher frequency non-harmonic frequency components.

11. 1. A computer-implemented method comprising: generating stored data indicating expected harmonic frequencies for a harmonic acoustic spectrum for each of a set of vocal chords; (a) for each audio chord of the set, spectrally analyzing a plurality of electronic waveforms resulting from one or more human subjects uttering the audio chord, the spectral analysis including, for each electronic waveform, estimating a fundamental audio frequency and identifying two or more fundamental or harmonic components, each having an intensity above a detection threshold and an audio frequency that is the fundamental audio frequency or a harmonic audio frequency equal to an integer multiple of the fundamental audio frequency; and (b) identifying as a primary cap frequency the highest harmonic acoustic frequency greater than 410 Hz among the identified harmonic components, and storing, for each vocal chord for which a primary cap frequency has been identified, an electronic indication of the primary cap frequency and the acoustic frequency of each identified fundamental or harmonic component in a tangible, non-transient, computer-readable storage medium that is not a transient, propagating signal; 4. A method, comprising using one or more electronic processors of a computer system programmed for:

12. 12. The method of claim 11, further comprising: for a subset of vocal chords having a common tonic, and for a plurality of utterances at a plurality of different fundamental frequencies by one or more human subjects, (i) estimating a common tonic focal frequency corresponding to the subset of vocal chords from the primary cap frequency; and (ii) storing an electronic indication of the tonic focal frequency in a tangible, non-transient computer-readable storage medium that is not a transient propagating signal.

13. 12. The method of claim 11, wherein the detection threshold varies as a function of acoustic frequency.

14. 12. The method of claim 11, wherein the stored data includes, for at least some of the set of sonic chords, harmonic frequencies of harmonic components of a primary band of a corresponding harmonic sound spectrum, the primary band including a primary cap frequency and one, two, three or more largest consecutive multiples of a fundamental sound frequency that are (i) less than the primary cap frequency, (ii) greater than 410 Hz, and (iii) greater than the smallest integer multiple of the fundamental sound frequency that is greater than 410 Hz and less than the primary cap frequency.

15. 12. The method of claim 11, wherein the stored data includes, for at least some of the audio chords of the set, harmonic frequencies of harmonic components of a secondary band of a corresponding harmonic audio spectrum, the secondary band including one or more harmonic components at a harmonic audio frequency greater than a smallest integer multiple of a fundamental audio frequency greater than 410 Hz and separated from a primary cap frequency by at least one intervening multiple of the fundamental audio frequency whose audio spectrum is devoid of harmonic components.

16. 12. The method of claim 11, wherein the stored data includes, for at least some of the set of vocal chords, harmonic frequencies of harmonic components of a base band of a corresponding harmonic acoustic spectrum, the base band including one or more harmonic components at one or more fundamental or harmonic acoustic frequencies less than 410 Hz or equal to the smallest integer multiple of the fundamental acoustic frequency between 410 Hz and a primary cap frequency.

17. (a) for each one of the additional harmonic or hybrid acoustic schemata, spectrally analyzing a plurality of electronic waveforms obtained when one or more human subjects speak that schema, wherein for each electronic waveform, the spectral analysis includes estimating a fundamental acoustic frequency and identifying two or more fundamental or harmonic components, each having an intensity above a detection threshold and a fundamental acoustic frequency or a harmonic acoustic frequency equal to an integer multiple of the fundamental acoustic frequency, each of the fundamental and harmonic acoustic frequencies being less than 410 Hz; 12. The method of claim 11, further comprising: (b) for each acoustic schema for which the plurality of waveforms are analyzed in part (a), storing, in a tangible, non-transient, computer-readable storage medium that is not a transient propagating signal, information electronically indicative of the presence or absence of one or more fundamental and harmonic acoustic frequencies for each identified fundamental or harmonic component, and higher frequency non-harmonic wave components.

18. 1. A computer-implemented method for synthesizing a time segment of an electronic waveform and producing a sound of an audio chord selected from a set of audio chords when the waveform segment is applied to an electro-acoustic transducer, comprising: (a) using the received, retrieved, or calculated data indicating a tonic note corresponding to a selected vocal chord and a focal frequency for that tonic note, determining a primary cap frequency using the focal frequency and the selected fundamental tone frequency, the primary cap frequency being (i) an integer multiple of the selected fundamental tone frequency, (ii) greater than 410 Hz, and (iii) closer to the focal frequency of the corresponding tonic note than to the focal frequency of any other vocal chord; and (b) including in the waveform segment a harmonic component at a primary cap frequency, the primary cap frequency being greater than the acoustic frequency of all other harmonics included in the waveform segment; using one or more electronic processors of a computer system programmed for:

19. 20. The method of claim 18, further comprising repeating the method of claim 18 for each voice chord in a temporal sequence of a plurality of different harmonic or hybrid segments, together with inharmonic or unvoiced segments and transition segments therebetween, that together constitute a human utterance.

20. 20. The method of claim 18, comprising applying waveform segments to an electro-acoustic transducer to produce sounds of a selected audio chord.

21. 20. The method of claim 18, further comprising including in the waveform segment, for one or more vocal chords of the set, a primary band of harmonic components, the primary band including a primary cap frequency and one, two, three or more largest consecutive multiples of a fundamental acoustic frequency that are (i) less than the primary cap frequency, (ii) greater than 410 Hz, or (iii) greater than the smallest integer multiple of the fundamental acoustic frequency that is greater than 410 Hz and less than the primary cap frequency.

22. 20. The method of claim 18, wherein for one or more vocal chords of the set, the waveform segments include harmonic components of a primary band, the primary band including harmonic components of a primary cap frequency and at least three highest consecutive multiples of a fundamental acoustic frequency that are (i) less than the primary cap frequency, (ii) greater than 410 Hz, and (iii) greater than a smallest integer multiple of the fundamental acoustic frequency that is greater than 410 Hz and less than the primary cap frequency.

23. For one or more vocal chords of said set, (a) using the received, retrieved, or calculated data indicating a secondary voice corresponding to a selected vocal chord and a focal frequency of the secondary voice, determining a secondary cap frequency using the focal frequency and the selected fundamental acoustic frequency, the secondary cap frequency being (i) an integer multiple of the selected fundamental frequency, (ii) greater than 410 Hz, and (iii) closer to the focal frequency of the corresponding secondary voice than to the focal frequency of any other voice; (b) the waveform segment includes one or more secondary band harmonic components, the secondary band including harmonic components at harmonic acoustic frequencies that are (i) greater than the smallest integer multiple of the fundamental acoustic frequency that is greater than 410 Hz and (ii) less than or equal to the secondary cap frequency; 20. The method of claim 18, further comprising:

24. 20. The method of claim 18, further comprising, for one or more vocal chords of the set, including in a waveform segment one or more harmonic components of a baseband, the baseband including harmonic components at one or more fundamental or harmonic acoustic frequencies (i) less than 410 Hz, or (ii) equal to the smallest integer multiple of the fundamental acoustic frequency that is between 410 Hz and a primary cap frequency.

25. 20. The method of claim 18, further comprising, for at least one additional time segment of the electronic waveform corresponding to a selected harmonic or hybrid acoustic schema from the set of acoustic schemata, including one or more reduced baseband-only harmonic components in the additional waveform segment, wherein the reduced baseband includes only harmonic components at one or more fundamental or harmonic acoustic frequencies below 410 Hz.

26. 20. The method of claim 18, further comprising, for at least one additional time segment of the electronic waveform corresponding to a selected harmonic or hybrid acoustic schema of the set of acoustic schemata, including in the additional waveform segment (i) one or more harmonic components of a reduced baseband only, the reduced baseband including only harmonic components of one or more fundamental or harmonic acoustic frequencies below 410 Hz, and (ii) one or more high frequency non-harmonic wave components.