Audio communication networks

• US 8,918,197 B2
• Filed: 06/13/2012
• Issued: 12/23/2014
• Est. Priority Date: 06/13/2012
• Status: Active Grant

• Current Assignee: Avraham Suhami
• Original Assignee: Avraham Suhami
• Inventors: Suhami, Avraham
• Primary Examiner(s): ELBIN, JESSE A
• Application Number:
  US13/495,648
• Publication Number:
  US 20130337796A1
• Time in Patent Office: 923 Days
• Field of Search: 700/94, 704/207, 704/223, 704/224, 704/271, 379/52, 379/88.01, 379/88.19, 379/93.26, 379/142.06, 379/390.02
• US Class Current: 700/94
• CPC Class Codes: G10L 2021/0135 Voice conversion or morphing
• CPC Class Codes: G10L 2021/065 Aids for the handicapped in understanding
• CPC Class Codes: G10L 21/003 Changing voice quality, e.g. pitch or formants
• CPC Class Codes: G10L 21/0364 for improving intelligibility
• CPC Class Codes: G10L 21/057 for improving intelligibility
• CPC Class Codes: H04M 3/569 using the instant speaker's algorithm speech detection per se G10L25/78
• CPC Class Codes: H04R 25/00 Deaf-aid sets , i.e. electro-acoustic or electro-mechanical hearing aids; Electric tinnitus maskers providing an auditory perception electrical stimulation of auditory nerves to promote the auditory function A61N1/36038; optical stimulation of auditory nerves to promote the auditory function A61N5/0622

Abstract

As the possible variations of “Hearing Thresholds”, “Hearing Loudness bandwidths” and “Voice Intonation” characteristics of people are finite, it is proposed to set a Database of these characteristics, where the data elements fully describe the Hearing and Talking characteristics of anyone, while many have the same characteristics. Thus any voice communication between two parties may be optimized by correcting the intensities of the call in the spectral domain, differently for each party and each ear. The optimizations are automatic given the “codes” of the parties and have a minimal latency. The system may be implemented either centrally in the world-wide-web or at the edges, in cellular phones, landline phones, VoIP, VoIM and in the audio parts of entertainment devices.

Claims

1. A database of “

audio communication codes”

for determining the optimal sound amplitudes to be relayed to the respective ears of the parties in an electronic audio communication, comprising “

hearing thresholds”

, “

hearing loudness” and

“

voice intonation”

tables in the spectral domain S, wherein the data elements of said tables may be related to one of the ears of the communicating person, and are enumerated in “

SPL dB”

units, wherein

an element (A_TL)_i of the “

hearing thresholds”

table represents the magnitude of amplification factors (A_TL) in the spectral domain, of sounds reaching one of the ears of a person, needed to bring his hearing level to a level, equivalent to that of a normal person, and,

elements (A_L)_j of the “

hearing loudness”

table represent the loudnesses extending from the minimal to the maximal tolerable loudness levels in the spectral domain (S) including the mid-level of loudness denoted as the “

comfort”

level L=C and,

an element (A_V)_k of the “

voice intonation”

table represents a distribution of the voice articulation loudness levels in the spectral domain (S) and wherein,

any combination of the elements [(A_T)_i, (A_L)_j, (A_V)_k] of said tables may be represented by a unique code in the spectral domain (S), wherein a “

multiplication”

of two Codes [(A_TL)_i, (A_L)_j, (A_V)_k] and [(A_TL)_l, (A_L)_m, (A_V)_n] each attributable to a different person, is not commutative, and is defined for a “

comfort”

loudness level L_i=C_i by

[(A_TC)_i,(A_LC)_j,(A_V)_k]*[(A_TC)_l,(A_LC)_m,(A_V)_n]=[(A_TC)_i+(A_LC)_j]*(A_V)_k[Σ

_S(A_V)_k/Σ

_S(A_TC+A_LC)_m] 

 

(I);

or

[(A_TC)_l,(A_LC)_m,(A_V)_n]*[(A_TC)_i,(A_LC)_j,(A_V)_k]=[(A_TC)_l+(A_LC)_m]*(A_V)_n[Σ

_S(A_V)_n/Σ

_S(A_TC+A_LC)_k] 

 

(II);

where

(I) gives the optimal sound level to be channeled to the ear of one party, and

(II) gives the optimal sound level to be channeled to the ear of the other communicating party.

2. A database of “

audio communication codes”

as in claim 1 wherein,

the ‘

distances’

, D_T, D_L, and D_V in SPLdB units, in the spectral domain, between the data elements (i) and the next level (i+1) in the 3 tables, defined as

D_T=Σ

_f(A_TL)_i−

(A_TL)_i+1;

D_L=Σ

_f(A_L)_i(A_L)_i+1;

D_V=Σ

_f(A_V)_i−

(A_V)_i+1

are of the same magnitude as are the indeterminacies in measuring said data in practice, and are of the order of ±

5-7 dB wherein,

any measured personal data may be approximated, for the purpose of calculating the optimal amplifications of the voice signals, by the predefined values of ‘

loudness in the spectral domain’

most close to it, and,

where f is the number of frequency bands.

3. A voice communication network storing a database of audio communication codes as defined in claim 1, where subscribers to the network are assigned a “

code”

representing the audio communication characteristics closest to their actual characteristics, for each of their ears and voice intonations, wherein,

elements of tables included in said codes are assigned to subscribers following one of an electronic Q&

A session and an audiogram performed in an audiology clinic, and electronically analysing the subscribers “

voice intonation”

characteristics.

4. A voice communication network as in claim 2 where the “

hearing threshold”

, the “

hearing “

loudness” and

the “

voice intonation”

characteristics in the frequency domain are self measured in a process, wherein,

the user listens to a preset list of complex tone bands, at increasing and decreasing loudness levels, and wherein,

the user'"'"'s perceived “

hearing threshold”

level and “

loudness bandwidth”

in the spectral domain are recorded by defining subjectively the hearing loudness levels that start from the “

don'"'"'t hear” and

“

barely hear”

levels to his “

comfortable”

level, up to the loud and excessively loud”

levels wherein,

the natural voice of the user articulating several phrases is recorded, thereafter letting the computer software analyze the power distribution in the frequency domain using a one of fourier transform, wavelet transform and a bank of filters, wherein,

the user'"'"'s measured ‘

hearing threshold’

, ‘

loudness levels’ and

‘

voice intonation’

characteristics are automatically compared with the lists of the predefined ‘

hearing threshold’

, ‘

loudness levels’ and

‘

voice intonation’

characteristic levels and,

 

the closest predefined characteristics, where the distances D_T, D_L, and D_V in SPLdB units, in the spectral domain, between the measured (m) and predefined (p) characteristics are less than [5N_f]dB, where N_f is the number of frequencies, are adopted and, if one or more of the measured characteristics are not close to one of the predefined ones, they are stored until such time that additional measurements support the fact that they constitute an additional set of characteristics, at which time they are added to the audio communication codes database.

5. Voice communication devices as in claim 4 for correcting outgoing calls addressed to people with hearing loss, comprising,

means for determining the individual hearing loss at both ears of the called person, as a function of his perceived loudnesses at various frequencies, by querying him on the phone and based on his responses, assigning the called person a “

code”

that fits his ‘

comfortable’

hearing characteristics, tuned to the caller'"'"'s voice intonation characteristics.

6. A networked audio communication device, storing a partial database of

“

audio communication codes”

as defined in claim 1, suitable to users of the network listed as “

contacts”

of said communications device, that have assigned “

codes”

,

representing their audio communication characteristics for each of their ears, and at least one equalizer, wherein

audio communications received by said device, and identified as a ‘

contact’

, have their audio characteristics relayed to the equalizer of said device, and the output of said equalizer transmitted to the speaker of the device and wherein,

any response to said “

contact”

is corrected according to the “

hearing”

characteristic of said contact as per his “

code” and

the ‘

voice intonation’

characteristic of the sender available in his ‘

code’

.

7. A database of audio communication codes as in claim 1, where the relative positions of its data elements in the frequency domain, each from the closest other element, representing the predefined characteristic features of the population, are gradually adjusted to better approximate the real-life self-measurements or clinical measurements of said characteristics performed by users of the database or available in medical archives.

8. A voice communication network as in claim 3 where the specific audio amplifications in the spectral domain, needed to optimize a voice communication between two parties possessing their respective ‘

Codes’

for each ear may be implemented by two preset equalizers for each ear, each equalizer for optimizing the voice call in one of the two directions and one of the ears.

9. A voice communication network storing a database of ‘

audio communication codes’

as defined in claim 1 wherein,

voice communications between subscribers of said network are executed by selectively modifying said voice data in the frequency domain, using equalizers located in a Bank of hardware equalizers, storing multiple copies of such preset equalizers equal to a multiple number of possible different characteristic variations, wherein,

each voice communication utilizes two preset equalizers, for the duration of the voice communication and by the end of said communication may be used for voice communication by different subscribers that have the same characteristics as the previous ones.

10. Voice communication means as in claim 1 including cellphones, landline telephones, RF transmitters, VoIP and VoIM applications wherein said means include:

hearing thresholds”

, “

hearing loudness” and

“

voice intonation”

tables of their users, and,

‘

hearing thresholds’

, ‘

hearing loudness’ and

‘

voice intonation tables of their contacts and,

one of hardware and software equalizers whose amplification factors may be set internally and,

microprocessors for setting the band amplification factors of said equalizers given the communicating parties identities and codes given the communicating parties automatic identification by “

Caller ID”

.

11. A voice communication network as in claim 10 wherein, the power spectrum in the spectral domain during intervals between syllables is identified as “

noise”

spectra and subsequently subtracted from the following speech spectra in the spectral domain.

12. A voice communication network as in claim 10 where syllable durations in transmitted speech are expanded on account of the time interval between said syllables.

13. A networked audio communication device, as in claim 10, comprising two equalizers, earphones and two focusing speakers wherein,

the received audio communication is ‘

corrected’

according to the appended ‘

code’

of the sender and the ‘

hearing profile’

of each of the ears of the user and wherein,

the two equalizers outputs are fed to two DACs and their outputs to one of two focusing speakers directed each to different ears and two earphones, and,

contains amplification means for amplifying the outputs of the equalizers separately.

14. Audio-Video entertainment devices incorporating hardware and software for implementing personal audio correction processes as in claim 1 wherein,

said entertainment devices comprise one of TV set, video monitor, radio set, audio/video projector, CD player, cellular phones, tablets, earphones, headphones and combinations thereof, and wherein,

said hardware and software comprise audio equalizers for programming in real time the spectral response of loudspeakers attached to or embedded in said entertainment devices, wherein,

said entertainment devices feature a wide spectral quality of the audio broadcast by the attached loudspeakers and wherein,

the specific programming of said equalizers reside in a database accessible using a personal code and may be uploaded by wireless, to said entertainment devices using one of a keypad, a computer and a cellphone, wherein,

said programming allows to optimize the spectral features of said entertainment devices depending on the audio sources, for implementing personalized hearing improvements.

15. One of a video and television monitors as in claim 14 whose left and right audio channels are relayed to two separate equalizers and respective directional loudspeakers, wherein,

said loudspeakers located within or outside the monitor box, transmit directional sound beams whose direction may physically be set and modified, and wherein, viewers viewers of said video and television monitors may upload by wireless, their personal codes specifying their hearing preferences and select the audio channel that implements said preferences.