Classifications

• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04S—STEREOPHONIC SYSTEMS 
  • H04S7/00—Indicating arrangements; Control arrangements, e.g. balance control
  • H04S7/30—Control circuits for electronic adaptation of the sound field
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04S—STEREOPHONIC SYSTEMS 
  • H04S5/00—Pseudo-stereo systems, e.g. in which additional channel signals are derived from monophonic signals by means of phase shifting, time delay or reverberation 
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
  • H04R5/00—Stereophonic arrangements
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04S—STEREOPHONIC SYSTEMS 
  • H04S7/00—Indicating arrangements; Control arrangements, e.g. balance control
  • H04S7/30—Control circuits for electronic adaptation of the sound field
  • H04S7/301—Automatic calibration of stereophonic sound system, e.g. with test microphone
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04S—STEREOPHONIC SYSTEMS 
  • H04S2420/00—Techniques used stereophonic systems covered by H04S but not provided for in its groups
  • H04S2420/01—Enhancing the perception of the sound image or of the spatial distribution using head related transfer functions [HRTF's] or equivalents thereof, e.g. interaural time difference [ITD] or interaural level difference [ILD]

Definitions

• the apparatus may include a speaker, a headphone (over-the-ear, on-ear, or in-ear), a microphone, a computer, a mobile device, a home theater receiver, a television, a Blu-ray (BD) player, a compact disc (CD) player, a digital media player, or the like.
• the apparatus may be configured to receive a virtualization profile including a digital audio filter with a design sample rate, resample the virtualization profile to a different sample rate, filter the audio signal with the resampled virtualization profile, and reproduce the filtered audio signal as sound.
• Various exemplary embodiments further relate to a method for processing an audio signal to influence the reproduction of the audio signal, the method comprising: sending a request to a server computer for a virtualization profile, wherein the request specifies a requested sample rate for the virtualization profile, and wherein the virtualization profile defines a digital audio filter; receiving from the server computer the virtualization profile with the requested sample rate; and filtering the audio signal based on at least the virtualization profile by performing a convolution of the audio signal with the virtualization profile with the requested sample rate.
• the virtualization profile represents an acoustic model of a production environment.
• the method further comprises causing the audio signal to be reproduced as sound through an audio transducer.
• Various exemplary embodiments further relate to a method for processing an audio signal to influence the reproduction of the audio signal, the method comprising: requesting a virtualization profile from a server computer, wherein the virtualization profile defines a digital audio filter; receiving from the server computer the requested virtualization profile with a design sample rate; resampling the virtualization profile at a required sample rate for the audio signal, responsive to a difference between the required sample rate and the design sample rate; and filtering the audio signal based on at least the virtualization profile with the required sample rate.
• resampling the virtualization profile comprises:
• filtering the audio signal comprises performing a convolution of the audio signal with the virtualization profile with the required sample rate.
• the method further comprises causing the audio signal to be reproduced as sound through an audio transducer simulating a production environment.
• Various exemplary embodiments further relate to a method for influencing reproductions of audio signals with virtualization profiles, the method comprising: storing a virtualization profile with a design sample rate, wherein the virtualization profile defines a digital audio filter; receiving a request for the virtualization profile from a client device, wherein the request specifies a requested sample rate for the virtualization profile;
• the digital audio filter represents an acoustic model of a production environment comprising at least one of a finite impulse response (FIR) filter, an infinite impulse response (IIR) filter, and a feedback delay network (FDN) filter.
• the virtualization profile causes the audio signal to be reproduced through an audio transducer simulating the production environment.
• the virtualization profile is stored as a series of filter coefficients in fixed point or float point values.
• resampling the virtualization profile comprises: interpolating the virtualization profile to obtain a representation of continuous-time bandlimited impulse response (CBIR); and resampling the CBIR at the requested sample rate.
• CBIR continuous-time bandlimited impulse response
• the method further comprises scaling the virtualization profile to a different sample rate to achieve a subjective audio effect.
• Various exemplary embodiments further relate to a non-transitory computer- readable storage medium storing computer-executable instructions that when executed cause one or more processors to perform operations comprising: storing a virtualization profile with a design sample rate, wherein the virtualization profile defines a digital audio filter; receiving a request for the virtualization profile from a client device, wherein the request specifies a requested sample rate for the virtualization profile; resampling the stored virtualization profile at the requested sample rate, responsive to a difference between the requested sample rate and the design sample rate; and transmitting the virtualization profile with the requested sample rate to the client device.
• the digital audio filter represents an acoustic model of a production environment comprising at least one of a finite impulse response (FIR) filter, an infinite impulse response (IIR) filter, and a feedback delay network (FDN) filter.
• the virtualization profile causes the audio signal to be reproduced through an audio transducer simulating the production environment.
• resampling the virtualization profile comprises: interpolating the virtualization profile to obtain a representation of continuous-time bandlimited impulse response (CBIR); and resampling the CBIR at the requested sample rate.
• CBIR continuous-time bandlimited impulse response
• Various exemplary embodiments further relate to an audio device for processing an audio signal, the audio device comprising: a communication interface configured for sending a request to a server computer for a virtualization profile, wherein the request specifies a requested sample rate for the virtualization profile, and wherein the virtualization profile defines a digital audio filter simulating a virtualized environment; and receiving from the server computer the requested virtualization profile with the requested sample rate; a storage device for storing the received virtualization profile; and a processor in
• Various exemplary embodiments further relate to an audio device for processing an audio signal, the audio device comprising: a communication interface configured for requesting a virtualization profile from a server computer, wherein the virtualization profile defines a digital audio filter simulating a virtualized environment; and receiving from the server computer the requested virtualization profile with a design sample rate; a storage device for storing the received virtualization profile; and a processor in communication with the storage device and the communication interface, the processor programmed for resampling the virtualization profile at a required sample rate for the audio signal, responsive to a difference between the required sample rate and the design sample rate; and filtering the audio signal based on at least the virtualization profile with the required sample rate.
• resampling the virtualization profile comprises:
• filtering the audio signal comprises performing a convolution of the audio signal with the virtualization profile at the required sample rate.
• FIG. 1 is a high-level block diagram illustrating an example environment 100 for cloud-based digital audio virtualization service, according to one embodiment.
• FIG. 2 is a block diagram illustrating components of an example computer system for cloud-based digital audio virtualization service, according to one embodiment
• FIG. 3 is a block diagram illustrating functional modules within a cloud server for the cloud-based digital audio virtualization service, according to one embodiment.
• FIG. 4A is a block diagram illustrating the bandlimiting effect of the CBIR resampling at a lower rate than the design sample rate, according to one embodiment.
• FIG. 4B is a block diagram illustrating the bandlimiting effect of the CBIR resampling at a higher rate than the design sample rate, according to one embodiment.
• FIG. 5 is a block diagram illustrating functional modules within a user device for the cloud-based digital audio virtualization service, according to one embodiment.
• FIG. 6 is a detailed interaction diagram illustrating an example process for providing cloud-based digital audio virtualization, according to one embodiment.
• a sound wave is a type of pressure wave caused by the vibration of an object that propagates through a compressible medium such as air.
• a sound wave periodically displaces matter in the medium (e.g. air) causing the matter to oscillate.
• the frequency of the sound wave describes the number of complete cycles within a period of time and is expressed in Hertz (Hz). Sound waves in the 12 Hz to 20,000 Hz frequency range are audible to humans.
• the present application concerns a method and apparatus for processing audio signals, which is to say signals representing physical sound. These signals may be represented by digital electronic signals.
• analog waveforms may be shown or discussed to illustrate the concepts; however, it should be understood that typical embodiments of the invention may operate in the context of a time series of digital bytes or words, said bytes or words forming a discrete approximation of an analog signal or (ultimately) a physical sound.
• the discrete, digital signal may correspond to a digital representation of a periodically sampled audio waveform.
• the waveform may be sampled at a rate at least sufficient to satisfy the Nyquist sampling theorem for the frequencies of interest.
• a uniform sampling rate of approximately 44.1kHz may be used.
• Higher sampling rates such as 96kHz or 192kHz may alternatively be used.
• the quantization scheme and bit resolution may be chosen to satisfy the requirements of a particular application, according to principles well known in the art.
• the techniques and apparatus of the invention typically would be applied interdependently in a number of channels. For example, it may be used in the context of a "surround" audio system (having more than two channels).
• the present invention may be implemented in a consumer electronics device, such as a Digital Video Disc (DVD) or Blu-ray Disc (BD) player, television (TV) tuner, Compact Disc (CD) player, handheld player, Internet audio/video device, a gaming console, a mobile phone, or the like.
• a consumer electronic device includes a Central Processing Unit (CPU) or Digital Signal Processor (DSP), which may represent one or more conventional types of such processors, such as an IBM PowerPC, Intel Pentium (x86) processors, and so forth.
• a Random Access Memory (RAM) temporarily stores results of the data processing operations performed by the CPU or DSP, and is interconnected thereto typically via a dedicated memory channel.
• the consumer electronic device may also include permanent storage devices such as a hard drive, which are also in communication with the CPU or DSP over an I/O bus. Other types of storage devices, such as tape drives and optical disk drives, may also be connected.
• a graphics card is also connected to the CPU via a video bus, and transmits signals representative of display data to the display monitor.
• External peripheral data input devices such as a keyboard or a mouse, may be connected to the audio reproduction system over a USB port.
• a USB controller translates data and instructions to and from the CPU for external peripherals connected to the USB port. Additional devices such as printers, microphones, speakers, and the like may be connected to the consumer electronic device.
• the present invention may have many different configurations and architectures. Any such configuration or architecture may be readily substituted without departing from the scope of the present invention.
• a person having ordinary skill in the art will recognize the above described sequences are the most commonly utilized in computer-readable mediums, but there are other existing sequences that may be substituted without departing from the scope of the present invention.
• Examples of the processor readable medium may include an electronic circuit, a semiconductor memory device, a read only memory (ROM), a flash memory, an erasable ROM (EROM), a floppy diskette, a compact disk (CD) ROM, an optical disk, a hard disk, a fiber optic medium, a radio frequency (RF) link, etc.
• the computer data signal includes any signal that may propagate over a transmission medium such as electronic network channels, optical fibers, air, electromagnetic, RF links, etc.
• the code segments may be downloaded via computer networks such as the Internet, Intranet, etc.
• the machine accessible medium may be embodied in an article of manufacture.
• the machine accessible medium may include data that, when accessed by a machine, may cause the machine to perform the operation described in the following.
• Embodiments of the present invention provide a system and a method for cloud- based digital audio virtualization.
• the method and system is organized around a cloud computing platform configured to aggregate, manage, and distribute virtualization profiles of the audio content.
• the virtualization profiles are generally derived from acoustic measurements of the production environment and uploaded to the cloud server.
• the user device may request a corresponding virtualization profile from the cloud server and apply the virtualization profile to the audio content to reproduce the audio content with desired production properties.
• the acoustic measurements for the virtualization profiles may be taken in rooms containing high fidelity audio equipment, for example, a mixing studio or a listening room.
• the room may include multiple loudspeakers, and the loudspeakers may be arranged in traditional speaker layouts, such as stereo, 5.1, 7.1, 11.1, or 22.2 layouts. Other non-standard or custom speaker layouts or arrays may also be used.
• Measurement room 110 shown in FIG. 1 contains a traditional 5.1 surround arrangement, including a left front loudspeaker, a right front loudspeaker, a center front loudspeaker, a left surround loudspeaker, a right surround loudspeaker, and a subwoofer. While a mixing studio having surround loudspeakers is provided as an example, the measurements may be taken in any desired location containing one or more loudspeakers.
• measurement server 112 may send one or more test signals to the one or more loudspeakers inside measurement room 110.
• the test signals may include a frequency sweep or chirp signal.
• a test signal may be a noise sequence such as a Golay code or a maximum length sequence.
• the acoustic measurements may be obtained by placing a measurement apparatus in an optimal listening position, such as a producer's chair.
• the measurement apparatus may be a free-standing microphone, binaural microphones placed within a dummy head, or binaural microphones placed within a test subject's ears.
• the measurement apparatus may record the audio signal received at the listening position and transfer the measurement data to server 112. From the recorded audio signals, measurement server 112 can generate a room measurement profile for each speaker location and each microphone of the measurement apparatus. Additional room measurements may be taken at other locations or orientations in the room, for example, at an "out of sweetspot" position. The "out of sweetspot" measurements may aid in determining the acoustics of measurement room 110 for listeners not in the optimal listening position or improving the acoustic models of the room space including the optimal listening position.
• the HRTF and BRIR filters may be digitized and stored as audio filter coefficients, and the room EQ can be represented by acoustic models that recreate the acoustics of the room. Similarly, the early and late reverberation can be digitized as audio filter coefficients or acoustic models for simulation. Both the HRTF or BRIR filters, room EQs and other acoustic models may be transmitted and/or stored as part of the virtualization profile.
• Measurement server 112 may process the virtualization profiles before uploading the virtualization profiles to cloud server 120.
• the processing of the virtualization profile by the measurement server 112 includes, but not limited to, validating, aggregating,
• a virtualization profile stored at cloud server 120 is measured by the measurement server 112 at a design sample rate, for example, 48 kHz. If a request from user devices 160 for the virtualization profile with a different requested sample rate than the designed sample rate, cloud server 120 may need to resample the virtualization profile in response to the request. Details of the resampling process are further described below in reference to FIG. 4. In alternate embodiments, the resampling of the virtualization profile can also be performed by the user devices 160 if so desired and indicated by the user devices 160 upon request.
• the audio content when an audio content starts to play on a user device, the audio content may include a flag in its bitstream indicating the user device about available virtualization profiles at cloud server 120 for download purchase.
• the user device may process the virtualization profile, for example, resample the virtualization profile to match the digital audio sample rate at the user device.
• the processed virtualization profile is then applied to filter the audio content for virtualized listening experience.
• an audio content may be processed in a mixing studio (e.g., measurement room 110), allowing the audio producer to measure the spatialized headphone mix that end-user hears.
• headphone 162 may download from cloud server 120 and store a virtualization profile generated by measurement server 112 from the measurement for the audio content. If headphone 162 applies the virtualization profile to the audio content, the acoustics and loudspeaker locations of the measured room will be recreated, and the audio content will sound similar to audio played back over the loudspeakers in the measured mixing studio.
• FIG. 2 is a block diagram illustrating components of an example computer able to read instructions from a computer-readable medium and execute them in a processor (or controller) to implement the disclosed system for cloud-based digital audio virtualization service.
• FIG. 2 shows a diagrammatic representation of a machine in the example form of a computer 200 within which instructions 235 (e.g., software) for causing the computer to perform any one or more of the methods discussed herein may be executed.
• the computer operates as a standalone device or connected (e.g., networked) to other computers.
• the computer may operate in the capacity of a server machine or a client machine in a server-client network environment, or as a peer machine in a peer-to-peer (or distributed) network environment.
• Processor 210 includes one or more central processing units (CPUs), graphics processing units (GPUs), digital signal processors (DSPs), application specific integrated circuits (ASICs), radio-frequency integrated circuits (RFICs), or any combination of these.
• Storage unit 230 comprises a non-transitory computer-readable storage medium 232, including a solid-state memory device, a hard drive, an optical disk, or a magnetic tape.
• the instructions 235 may also reside, completely or at least partially, within memory 220 or within processor 210's cache memory during execution thereof by computer 200, memory 220 and processor 210 also constituting computer-readable storage media. Instructions 235 may be transmitted or received over network 140 via network interface 260.
• Input devices 250 include a keyboard, mouse, track ball, or other type of alphanumeric and pointing devices that can be used to input data into computer 200.
• the graphics adapter 212 displays images and other information on one or more display devices, such as monitors and projectors (not shown).
• the network adapter 260 couples the computer 200 to a network, for example, network 140. Some embodiments of the computer 200 have different and/or other components than those shown in FIG. 2.
• the types of computer 200 can vary depending upon the embodiment and the desired processing power.
• the term "computer” shall also be taken to include any collection of computers that individually or jointly execute instructions 235 to perform any one or more of the methods discussed herein.
• Cloud server 120 stores virtuahzation profiles uploaded by measurement server 112, and distribute the virtuahzation profiles to user devices 160 over network 140.
• FIG. 3 is a block diagram illustrating functional modules within a cloud server 120 for the cloud-based digital audio virtuahzation service.
• cloud server 120 comprises a profile manager 310, a profile database 320, a profile-processing module 330, and a network interface 340.
• module refers to a hardware and/or software unit used to provide one or more specified functionalities.
• a module can be implemented in hardware, software or firmware, or a combination of thereof.
• Other embodiments of could server 120 may include different and/or fewer or more modules.
• the profile manager 310 receives measurement data 305 uploaded by the measurement server 112.
• the measurement data 305 may include raw room measurements and/or virtuahzation profiles processed by the measurement server 112.
• the profile manager 310 may also validate, encode, enciypt and compress the measurement data 305 before storing the processed measurement data 305 in the profile database 320.
• a room response measurement e.g., room 110
• A/D analog-to-digital converter
• the resulting filter coefficients are then stored at the profile database 320 as a virtualization profile associated with the measurement room 110.
• a virtualization profile When stored at profile database 320, a virtualization profile may be indexed and retrieved by a unique identifier, such as an MD5 checksum or any hash values generated by other hash functions.
• the unique identifiers of the virtualization profiles can also be derived from other identifying information, such as measurement room information, measured loudspeaker layout information, measurement location information, measurement equipment information, associated audio content identifiers, and/or licensing and ownership information.
• the virtualization profile identifiers can be generated by the profile manager 310 or received from the measurement server 112.
• the profile database 320 may also store other audio production profiles, such as playback device profiles and listener hearing profiles.
• a profile request 335 may include identification information of the requested virtualization profile.
• the request 335 can specify a unique profile identifier, or other identification information such as associated audio content, measurement rooms, and/or production or license owners.
• the profile request 335 may further specify parameters, such as sample rate, A/D length, and bitrate, of the virtualization profile requested.
• the request 335 may be generated by the user devices 160 automatically based on preconfigured user device profiles and/or user preferences.
• the network interface 340 can provide a graphic user interface (GUI), such as a webpage, to the user devices 160 and prompt users to fill in identification information or other parameters of the requested virtualization profiles in advance or on the fly.
• GUI graphic user interface
• the network interface 340 passes requests for virtualization profiles to the profile manager 310, which searches the requested profiles from the profile database 320. Search results are then forwarded to the profile-processing module 330.
• a search result may include more than one virtualization profiles, for example, a list of profiles of multiple rooms' measurements.
• the profile-processing module 330 may choose one or more profiles from the search results and pass the resulting virtualization profiles to the network interface 340, which transmits the virtualization profiles as a response 345 to the profile request 335 back to the requesting user device 160.
• the profile database may also log types of virtualization profiles, number of requests, among other preference data. This may allow the cloud server to provide customized recommendations to users based on usage history.
• the profile-processing module 330 helps select which room's acoustics should be returned to the requesting user. For instance, the user may prefer audio content to be processed with a virtualization profile that is most similar to the acoustics of his or her current room. In this case, the profile-processing module 330 may need to communicate with the client devices 160 for acoustic measurement of the user's room with one or more tests. For example, the user may clap his or her hands in the current room, and the hand clap is recorded and processed to determine the acoustic parameters of the room. Alternatively or in addition, other environmental sounds, such as speech, may be analyzed. The tests can be processed either by the cloud server 120 or at the client devices 160. In alternate embodiments, the profile-processing module 330 may simply respond to the user request with one or more virtualization profiles and let the requesting user to select.
• users may request virtualization profiles or filters with a different sample rate than the design sample rate captured by the measurement server 112 and/or stored at the profile database 320.
• One solution is to design the filters, such as an infinite impulse response (IIR) Butterworth filter, with simple operations to be automated on the fly.
• IIR infinite impulse response
• Such "simple" designs often require operations (e.g., sin/cos or log) not consistent or available with high enough precision across all platforms. This lack of precision is further compounded when coupling with fixed-point systems.
• the cloud- based virtualization filters e.g., HRTFs, BRIRs, and Room EQs
• Another option is to design filters at all possible sample rates offline and store the filters in a database by predicting and measuring room response for every sample rate that might be needed. This method may consume a significant amount of memory and storage because multiple filters are needed and each filter with a number of sample rates, making it unacceptable especially for embedded systems.
• a third option is to distribute both audio content and filters at the same sample rate. However, it is impractical to require numerous audio applications across various software platforms to process digital audio and filters all at a specific sample rate. Fixing audio sample rate may also cause portability problem and licensing issues. A requirement for end users to clock their global audio path to a fixed sample rate may be prohibitive in terms of computing resources, such as CPU power, memory consumption, and battery life, among other bill of materials cost.
• the preferred solution is to design filters containing spectral resolution suitable for any sample rate and automatically adapted to any playback rate on the fly. It is the well- known that a sampled signal is bandlimited to half of the sampling rate (i.e., the Nyquist frequency). Shannon's sampling theorem also suggests that the original signal can be exactly and uniquely reconstructed by interpolating between the sampled values if the sampling rate is higher than the Nyquist frequency. Therefore, the method of bandlimited interpolation, which operates on the foundation Nyquist- Shannon sampling theorem, provides a means of reproducing a continuous-time, yet bandlimited impulse responses from room measurements, rather than "filter coefficients" only at a design sample rate.
• the profile-processing module 330 resamples the
• virtualization profile involves applying interpolation on the virtualization profile, such as HRTF and BRIR filters, to obtain a continuous bandlimit impulse response (CBIR).
• CBIR continuous bandlimit impulse response
• the interpolated CBIR is then resampled at the requested sample rate before transmitting to the user devices 160.
• the interpolated CBIRs and resampled virtualization profiles can be stored and/or cached by cloud server 120 for further use if storage space allows.
• the method allows filters to be designed once, and later adjusted to any requested sample rates without dependency on any special functions that might deviate due to different
• the bandlimited interpolation fits perfectly to the audio filter design choice because the audio frequencies of interest lie in the audible range of 20Hz-20kHz.
• these filter taps can be interpolated and resampled at any rate to cover the spectrum of interest. For instance, if the interpolated CBIR is resampled at a rate higher than the design sample rate, an input audio signal passing the CBIR is automatically
• bandlimited at the original Nyquist frequency of the filter (i.e. 20kHz). Whereas in case of a lower resample rate, bandlimited interpolation becomes effectively a low-pass filter for the CBIR with a cutoff frequency at Nyquist frequency of the lower resample rate. In the latter case, loss of the filter specification for higher frequencies is acceptable because those frequencies are absent from the input audio signal being processed in the first place.
• FIG. 4A is a diagram illustrating the bandlimiting effect of the CBIR resampling at a lower rate than the design sample rate, according to one embodiment.
• the impulse response prototype 400 is a virtualization profile stored at the profile database 320 with a design sample rate of R ⁇ .
• An audio input 401 has a sample rate of R/. where R / ⁇ 3 ⁇ 4.
• the profile-processing module 330 first interpolates 405 the impulse response prototype 400 to obtain an CBIR 410, which is subsequently resampled 407 to produce an impulse response 420 with a target sample rate of i? / .
• Audio output 422 is the result of filtering the audio input 401 through the resampled impulse response 420.
• the audio output 422 has a narrower bandwidth compared to the ideal audio output 412, which is filtered by the CBIR.
• This process demonstrates a finite impulse response (FIR) design technique based on sampling the continuous impulse response of the ideal filter represented by the interpolated CBIR.
• FIG. 4B is a diagram illustrating the bandlimiting effect of the CBIR resampling at a higher rate than the design sample rate, according to one embodiment.
• audio input 402 has a sample rate of R h .
• the profile-processing module 330 resamples 409 the interpolated CBIR 410 to produce an impulse response 430 with a target sample rate of 3 ⁇ 4.
• Audio output 432 is the result of filtering the audio input 402 through the resampled impulse response 430.
• the resulting audio output 422 is bandlimited at the original Nyquist frequency of the original impulse response.
• the bandlimited interpolation is suitable for resampling audio particularly because it reconstructs signal sampled at given points rather than approximating the signal through or around the sample points.
• a moving sine function is utilized for interpolating the impulse response prototype; the sine function serves as the band-limiting low pass filter.
• a special windowed sine function is used instead. Generally this is achieved with a Kaiser window to control the trade-off between the stop-band attenuation and pass-band transition width.
• the profile-processing module 330 also provides controlled frequency scaling.
• a scaling of a high-frequency resonance can render HRTFs a personalization effect roughly associated with ear-size and/or shape.
• the sample rate of a filter is adjusted by a factor relative to the true audio sampling rate.
• a factor of 1 i.e. equal to the true audio rate
• a factor less than 1 scales spectral features higher in frequency
• a factor greater than 1 scales spectral features lower in frequency.
• the frequency scaling reflects compression or expansion of the impulse response in time domain caused by the difference between the sample rates of the filter and the signal.
• the headphone 162 communicates with the cloud server 120 via the network interface 510, which may be wired or wireless.
• the headphone 162 may be associated with a unique user account for the audio virtualization service at the cloud server 120.
• the user account may include information about the user of the headphone 162, such as the user's identification information, the user's hearing profiles and/or playback device profiles, and other user preferences.
• the network interface 510 forwards a user request 325 for virtualization profiles to the cloud server 120 and receives response 345 including one or more virtualization or room measurement profiles from the cloud server 120.
• the virtualization profiles received by the network interface 510 can be associated with the user account and passed to the profile memory 520 for use and storage.
• the virtualization profiles may be transmitted to the headphone 162 embedded in metadata of the audio content or separately from the audio content.
• the downloaded virtualization profile includes the resampled room measurement profiles, such as HRTF filter coefficients resampled to match the sample rate of the audio content.
• the interpolation and resampling of the virtualization profiles is performed by the cloud server 120 at the request of the headphone 162 indicating the target sample rate of the audio content.
• the profile memory 520 then passes the virtualization profiles to the audio processor 540, which processes the audio content by performing a direct convolution of the audio content with the downloaded virtualization profiles.
• the profile-processing module 530 may create an acoustic model of the measurement room 110 and forward the acoustic model to the audio processor to process the audio content.
• the early room response parameters and the late reverberation parameters may be convolved with the audio content by the audio processor 540.
• the headphone 162 or any other user devices may request original virtualization profiles, such as H TF filter coefficients, at the design sample rates from the cloud server 120 without any processing.
• the profile- processing module 530 can process the virtualization profile locally. For example, if the design sample rates of the original virtualization profiles are different from the target sample rates of the audio content, the profile processing module 530 first needs to perform interpolation on the original HRTF and BRIR filters, to obtain a continuous bandlimit impulse response (CBIR). The interpolated CBIR is then resampled at the target sample rate before passing to the audio processor.
• the resampled filter coefficients can also be stored and/or cached by the profile memory 520 if necessary.
• the profile-processing module 530 and the audio processor 540 may process the virtualization profiles and the audio content at the time of playback and/or prior to the time of playback.
• the processing of the audio content and the virtualization profiles may be distributed to other user devices.
• the audio content may be pre-processed with some virtualization profiles at a local server and transmitted to headphone 162.
• the cloud-based virtualization may be constructed in such a way as to allow pre-processing of audio by content producers. This process may generate an optimized audio track designed to enhance user device playback in a manner specified by the content producer or to retain the desired attributes of the originally mixed surround soundtrack that provides the listener the sonic experience in the original studio.
• the result of audio processing by the profile-processing module 530 and the audio processor 540 may be a bit stream that can be decoded using any audio decoder.
• the bit stream may include a flag that indicates whether or not the audio has been processed with the virtualization profiles. If the bit steam is played back using a legacy decoder that does not recognize the flag, the content may still be played, but without showing any indication on the virtualization of the audio content.
• FIG. 6 is a detailed interaction diagram illustrating an example process for providing cloud-based digital audio virtualization, according to one embodiment. It should be noted that FIG. 6 only demonstrates one of many ways in which the embodiments of the cloud-based virtualization may be implemented.
• the method for providing the digital audio virtualization service involves the measurement server 112, the cloud server 120, and the user devices 160. The method begins with the measurement server 112 capturing 610
• the measurement server 112 Based on the room measurements, the measurement server 112 generates 612 virtualization profiles, such as head-related transfer function (HRTF) or binaural room impulse response (BRIR) filters, and then uploads 614 the virtualization profiles to the cloud server 120.
• HRTF head-related transfer function
• BRIR binaural room impulse response
• the cloud server 120 may process 620 the virtualization profiles uploaded by the measurement server 112. For example, the cloud server 120 may validate, encode, encrypt, and compress the virtualization profiles before storing 622 them in its database (e.g., profile database 320).
• the virtualization profiles such as HRTF and BRIR filters, are often stored as filter coefficients sampled at a design sample rate. As described above, it is preferable to design the filter at the highest possible sample rate, which fits the filter storage block length for the purpose of interpolation and resampling. For example, a virtualization filter can be sampled as high as 192kHz with 64 bit A/D converter length.
• the user devices 160 may also directly request 640 the original virtualization profiles 640 or without specifying any sample rate from the cloud server 160, which will respond by transmitting 642 the requested virtualization profile with design sample rate (without any resampling).
• the user devices 160 can resample 644 the virtualization filter of the profile to the required sample rate of the audio content before filtering 646 the audio content with at least the virtualization profile for a playback with environment virtualization effect.
• the resampling operation performed at the user devices 160 may include a direct convolution of the audio content with the resampled virtualization filters of the profiles.
• the method and apparatus disclosed in embodiments deliver a flexible filter design that adapts to variable sample rates for influencing the reproduction of the audio signal.
• the digital filters stored in the cloud servers are designed with the highest possible sample rate for the purpose of interpolation and resampling to different target sample rates.
• User devices when plays back digital audio content, may directly download those digital filters resampled to proper sample rates by the cloud server, or acquire and resample the digital filters locally for virtualized listening experience.

Landscapes

• Physics & Mathematics (AREA)
• Engineering & Computer Science (AREA)
• Acoustics & Sound (AREA)
• Signal Processing (AREA)
• Stereophonic System (AREA)
• Complex Calculations (AREA)
• Circuit For Audible Band Transducer (AREA)
• Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)

Priority Applications (4)

Applications Claiming Priority (2)

Publications (1)

Family

ID=55631222

Family Applications (1)

Country Status (6)

Cited By (1)

Families Citing this family (18)

Citations (6)

Family Cites Families (18)

• 2014
  • 2014-10-03 US US14/506,187 patent/US9560465B2/en active Active
• 2015
  • 2015-06-30 WO PCT/US2015/038635 patent/WO2016053432A1/en not_active Ceased
  • 2015-06-30 CN CN201580060946.XA patent/CN107251009B/zh active Active
  • 2015-06-30 EP EP15846638.3A patent/EP3201791A4/en not_active Ceased
  • 2015-06-30 KR KR1020177011766A patent/KR102502465B1/ko active Active
  • 2015-06-30 JP JP2017518126A patent/JP6640204B2/ja active Active

Patent Citations (6)

Non-Patent Citations (1)

Also Published As

Similar Documents

Legal Events