DE102013211283A1 - Playback of audio data using distributed electroacoustic transducers in networked mobile devices - Google Patents

Abstract

The present invention provides a method of reproducing a sound field emanating from a virtual source using a plurality of mobile terminals each having at least one electroacoustic transducer in a local space, the terminals being connected by a data network. The respective positions of the terminals in the local space are determined, on the basis of a signal transmitted via the data network, driving functions for the electroacoustic transducers for reproducing the sound field are determined depending on the position of the virtual source and the determined positions of the terminals in the local space using the thus determined driving functions, the sound field is reproduced by the electroacoustic transducers in the terminals. The invention further provides a method for full duplex communication and an apparatus for performing these methods.

Description

The present invention relates to the synthesis of a sound field for spatial audio reproduction by means of several networked terminals, which are equipped with speakers and preferably microphones, and in particular their use in the full-duplex, hands-free communication.

In general, in multichannel audio playback, the loudspeakers are controlled such that a spatial hearing impression is generated in a predefined range. Here, numerous methods for audio playback or for the physical synthesis of a sound field are known. By way of example, stereophony, wavefield synthesis (see, for example, US Pat. AJ Berkhout, D. de Vries, and P. Vogel. Acoustic control by wave field synthesis. Journal of the Acoustic Society of America, Vol. 93 (5): 2764-2778, May 1993 ) or higher-order Ambisonics ( J. Daniel, Representation of Acoustics Acoustics, Application for the Transmission and Playback of Matters in Multimedia, PhD thesis, Université Paris 6, 2001 ) to be named. These methods of multi-channel playback methods assume fixed predefined speaker positions.

Methods for echo cancellation and echo cancellation and for signal improvement, in particular in full-duplex communication using hands-free devices, are described, for example, in US Pat E. Hänsler, G. Schmidt, Topics in acoustic echo and noise control: selected methods for the cancellation of acoustical echoes, the reduction of background noise, and speech processing, Springer-Verlag, Berlin 2006 described.

The present invention provides an improved method and a device for reproducing a sound field, which can advantageously be used in particular in full-duplex communication. According to the invention, a plurality of mobile, networked terminals, in particular smartphones, are used for sound field synthesis.

The present invention is defined by the independent claims. The dependent claims define embodiments of the invention.

The present invention provides a method of reproducing a sound field emanating from a virtual source using a plurality of mobile terminals each having an electroacoustic transducer in a local space, the terminals being connected by a data network. First, the respective positions of the terminals in the local space are determined. On the basis of a signal transmitted over the data network, driving functions for the electroacoustic transducers for reproducing the sound field are determined as a function of the position of the virtual source and the determined positions of the terminals in the local space. Using the thus determined driving functions, the sound field is reproduced by the electro-acoustic transducers in the terminals.

The terminals may be interconnected by a star-shaped network with an external or a local server, or by a fully meshed network. The network is preferably a radio network, in particular a mobile radio network such as GSM, GPRS, UMTS or LTE, WLAN or Bluetooth.

To determine the position of the terminals, a location method may be used, for example a video-based location method, an audio-based location method and / or a location method using GPS. Additionally or alternatively, the position of the terminals may also be determined by manual input by the user.

To compensate for the delay caused by the transmission of the signal, playback is preferably synchronized by the individual terminals.

The content of the sound field to be reproduced can be selected by one of the terminals. This content can be represented in particular by the signal transmitted via the data network. This is the case in particular if the content of the sound field to be reproduced is the content of an audio signal from a distant room, for example the audio signal during a telephone call (eg by means of GSM or VoIP). On the other hand, a content that is present in all terminals, for example in the form of a stored audio file, can also be reproduced. Then a time hand can be sent to the audio file over the data network.

For the inventive method for full-duplex communication between a remote room and a local room, the method described above is used to reproduce the audio signal in the form of a sound field emanating from a virtual source. Furthermore, the plurality of mobile terminals record the acoustic signal to be transmitted from the local room to the distant room. The local server, the external server and / or the individual mobile terminals preferably have a device for echo compensation.

The invention further provides an apparatus for carrying out the method according to the invention, wherein the apparatus comprises a plurality of mobile terminals, for example smartphones, each having an electro-acoustic converter, which are connected via a data network.

The invention will be described in more detail below with reference to embodiments with reference to the accompanying figures.

1 shows various network topologies used in accordance with embodiments of the present invention to which the mobile terminals may be connected, namely (a) a star-shaped network with an external server, (b) a star-shaped network with a local server, and (c) a fully-meshed network.

2 shows an example of a user interface on a mobile terminal in an active session using a method according to an embodiment of the invention.

• 1A. Ein sternförmiges Netzwerk, worin alle lokalen Teilnehmer über Ihre Smartphones über einen zentralen Hauptknoten verbunden sind. Der Verbindungsaufbau zum Server erfolgt über Einwählen in einen zentralen Dienst (siehe 1a).
• 1B. Ein sternförmiges Netzwerk, worin alle lokalen Teilnehmer über Ihre Smartphones über einen lokalen Hauptknoten verbunden sind. Hier dient beispielsweise eines der Smartphone als Hotspot (siehe 1b).
• 2. Ein vollvermaschtes Netzwerk, in dem alle lokalen Teilnehmer miteinander über ein lokales bidirektionales Netzwerk, beispielsweise über ein Bluetooth-Netzwerk verbunden sind (siehe 1c).

The audio playback according to an embodiment of the present invention will be described below in detail with reference to the flow of a typical session, where a session may be, for example, a telephone conference or the playback of a predetermined audio file transmitted over the network or stored on all used terminals (smartphones) In principle, in such a session, the process is based on two network topologies, whereby a topology can be subdivided into two types depending on the server location:

• 1A. A star-shaped network where all local subscribers are connected via their smartphones via a central hub. The connection to the server is made by dialing into a central service (see 1a ).
• 1B. A star-shaped network where all local subscribers are connected via their smartphones via a local master node. For example, one of the smartphones serves as a hotspot here (see 1b ).
• 2. A fully meshed network in which all local participants are connected to each other via a local bi-directional network, for example via a Bluetooth network (see 1c ).

A session can be started from any terminal. To register at a meeting, each participant is assigned a recognizable identification number (ID). After logging in, the participants submit their current positions, which they can calculate using common localization peripherals (eg GPS). The positioning can also be done manually or corrected by the users sequentially entering their relative positions. For manual positioning, for example, the users on the display may be shown a blank map of neighboring cells in which the users select their cell in which they are located.

Another type of localization can be done by all users placing their smartphones with the display facing up on a surface (for example, a meeting room table), except for a user who then calls the calibration function. In the calibration function, the subscribers' displays display various patterns and / or colors that are captured and evaluated by the user's camera that started the calibration function. By means of standard pattern recognition methods, the positions of the participating smartphones are thus determined and sent centrally to all subscribers.

In addition to this possibility of video-based localization, localization based on audio data can also be performed. For this purpose, the participants' devices are requested to play predefined different sound sequences. These are recorded with the microphones of the terminal on which the calibration function is performed and processed by means of standard methods of audio localization locally or on an external server. A suitable method for this purpose is, for example, in H. Buchner, R. Aichner, and W. Kellermann, "TRINICON-based Blind System Identification with Application to Multiple-Source Localization and Separation," in Blind Speech Separation, S. Makino, H. Sawada, and T.-W. Lee, ed. Springer Netherlands, 2007, pp. 101-147 described.

To improve the position determination, a combination of the presented localization methods can also be used.

During a session, an active map is displayed with the participants, as shown schematically in 2 is shown. The user interface should preferably have a feature that allows virtual sources to be added. Such a virtual source can either be a telephone number dialed to start a telephone conference. However, it can also be a recorded audio signal (eg a .wav, .mp3 file, etc.), which is either stored on all participating terminals or transmitted via the radio network.

• – Bestimmung der Ansteuerungsfunktion der Lautsprecher in Abhängigkeit von der gewünschten Position der virtuellen Quelle und der ermittelten geometrischen Anordnung. Es wird vorzugsweise zu jeder Quelle ein Satz von Lautsprechen (Smartphones) selektiert, die bei der Wiedergabe dieser virtuellen Quelle aktiv werden sollen.
• – Für die oben genannten zwei Fälle der Wiedergabe ist Folgendes zu beachten: – Inhalt nur an einem Gerät gespeichert: In diesem Fall wird jedem Teilnehmer, der entsprechend der Selektion bei der Wiedergabe einer bestimmten Quelle aktiv sein soll, der Inhalt zugesandt. – Alle Teilnehmer verfügen über den gesamten (musikalischen) wiederzugebenden Inhalt: Hier wird lediglich der Zeitzeiger auf die aktive Datei ermittelt.
• – Synchronisierung der Endgeräte. Es muss sichergestellt werden, dass die durch die Funkübertragung verursachte Verzögerung der Daten kompensiert wird.
• – Schnelle Faltung des wiederzugebenden Inhalts mit den eigenen Ansteuerungsfunktionen.

To reproduce by synthesizing the desired sound field, the following steps are performed:

• - Determining the driving function of the speakers depending on the desired position of the virtual source and the determined geometric arrangement. Preferably, a set of voices (smartphones) are selected for each source to be active when playing this virtual source.
• - For the above mentioned two cases of playback, the following should be noted: - Content stored on one device only: In this case, the content will be sent to each participant that is to be active according to the selection when playing a particular source. - All participants have the entire (musical) content to be played back: here only the time hand is determined on the active file.
• - Synchronization of the terminals. It must be ensured that the data delay caused by the radio transmission is compensated.
• - Fast convolution of the content to be played with its own control functions.

• 1A. Die lokalen Teilnehmer können einen Teilnehmer in einem fernen Ende über den externen Hauptknoten anwählen. In diesem Fall wird vorzugsweise auf dem Server ein Echokompensator (engl. Acoustic Echo Canceler, AEC – siehe beispielsweise E. Hänsler, G. Schmidt, Topics in acoustic echo and noise control: selected methods for the cancellation of acoustical echoes, the reduction of background noise, and speech processing, Springer-Verlag, Berlin 2006 oder H. Buchner, J. Benesty, und W. Kellermann, Generalized multichannel frequency-domain adaptive filtering: efficient realization and application to hands-free speech communication, Signal Processing, Bd. 85, Nr. 3, S. 549–570, 2005 ) oder Echounterdrücker (engl. Acoustic Echo Suppressor, AES – siehe beispielsweise C. Faller and C. Tournery, Stereo acoustic echo control using a simplified echo path model, in Proc. IWAENC, 2006 oder EP 1 715 669 A1 ) implementiert, so dass der Server nur echofreies Signal vom nahen ins fernen Ende übermittelt.
• 1B. Nur der Teilnehmer, dessen Smartphone oder Endgerät als Server dient, kann einen Teilnehmer im fernen Ende anrufen. Dann wird der Echounterdrücker (bzw. -kompensator) vorzugsweise auf diesem Gerät implementiert sein.
• 2. Jeder Teilnehmer kann selbst über den eigenen Mobilfunkanbieter einen Teilnehmer in einem fernen Ende anrufen und das Sprachsignal dieses im nahen Ende je nach gewünschter Position der virtuellen Quelle über dem lokalen Netzwerk weiterreichen (ausstrahlen). In diesem Szenario sollte der Echounterdrücker (bzw. -kompensator) auf jedem der beteiligten Endgeräte implementiert sein.

For a conference circuit for full-duplex communication, a distinction is made below between the playback and recording techniques:
Again, depending on the network topology, a distinction is made between three cases:

• 1A. The local subscribers can dial a subscriber in a far end via the external main node. In this case, preferably on the server an echo canceller (Acoustic Echo Canceler, AEC - see, for example E. Hänsler, G. Schmidt, Topics in acoustic echo and noise control: selected methods for the cancellation of acoustical echoes, the reduction of background noise, and speech processing, Springer-Verlag, Berlin 2006 or H. Buchner, J. Benesty, and W. Kellermann, Generalized multichannel frequency-domain adaptive filtering: efficient realization and application to hands-free speech communication, Signal Processing, Vol. 85, No. 3, pp. 549-570, 2005 ) or echo suppressors (Acoustic Echo Suppressor, AES - see for example C. Faller and C. Tournery, Stereo acoustic echo control using a simplified echo path model, in Proc. IWAENC, 2006 or EP 1 715 669 A1 ) so that the server transmits only echo-free signal from the near to the far end.
• 1B. Only the subscriber whose smartphone or terminal serves as a server can call a subscriber in the far end. Then, the echo canceller (or compensator) will preferably be implemented on this device.
• 2. Each subscriber can even call a subscriber in a far end via their own mobile service provider and pass on the voice signal in the near end depending on the desired position of the virtual source on the local network (broadcast). In this scenario, the echo suppressor (or compensator) should be implemented on each of the participating terminals.

Just as on the playback side, several electroacoustic transducers are also available on the recording side due to the terminals distributed in the local area. In principle, the multi-channel recording can also be used for spatial processing, in particular for the suppression of interference signals. When recording, the invention therefore sees the application of already known multi-channel adaptive methods for blind source separation (see, for example, US Pat. H. Buchner, R. Aichner, and W. Kellermann, "TRINICON-based Blind System Identification with Application to Multiple-Source Localization and Separation," in Blind Speech Separation, S. Makino, H. Sawada, and T.-W. Lee, ed. Springer Netherlands, 2007, pp. 101-147 ) and interference suppression (beamforming, described for example in Brandstein and D. Ward, Microphone arrays: signal processing techniques and applications, Birkhäuser 2001 ), single- and multichannel noise suppression, as also described in the above-mentioned publications by Brandstein and D. Ward and by E. Hänsler and G. Schmidt), the already determined positions of the terminals (see above) as prior information about the Position of the microphones can serve. Furthermore, a simple selection of the microphones to be used during recording is possible, for. Based on estimates of signal-to-noise power ratios.

The invention also provides for a possible combination of said signal enhancement methods. Thus, for example, a hierarchical approach can be followed, in which initially a cluster of participants can be formed by simple selection and then a blind source separation is performed in this cluster.

In the processing of the audio data, a parallelization strategy is preferably pursued on the server or terminal serving as server because of the possible multiplicity of channels for complexity reasons. For example, processing in the transformation domain (see H. Buchner and S. Spors, A General Derivation of Wave-Domain Adaptive Filtering and Application to Acoustic Echo Cancellation, Proc. Asilomar Conference at Signals, Systems, and Computers, Pacific Grove, CA, USA, Oct. 2008 or K. Helwani, H. Buchner, and S. Spors, Source-Domain Adaptive Filtering for MIMO Systems with Application to Acoustic Echo Cancellation, in the IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2010 ), MapReduce (see J. Dean and S. Ghemawat, " MapReduce: simplified data processing at large clusters ", 6th Symposium on Operating Systems Design and Implementation, pp. 107-113, 2004 ), or a combination of these are used.

While the invention has been illustrated and described in detail by the figures and the accompanying description, this description and detailed description are to be considered illustrative and exemplary and not limiting as to the invention. It is understood that those skilled in the art can make changes and modifications without departing from the scope of the following claims. In particular, the invention also includes embodiments with any combination of features that are mentioned or shown above in various aspects and / or embodiments.

The invention also includes individual features in the figures, even if they are shown there in connection with other features and / or not mentioned above.

Furthermore, the term "comprising" and derivatives thereof does not exclude other elements or steps. Also, the indefinite article "a" and "derivatives" and derivatives thereof do not exclude a variety. The functions of several features listed in the claims may be fulfilled by one unit. The terms "substantially", "approximately", "approximately" and the like in connection with a property or a value in particular also define precisely the property or exactly the value.

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• EP 1715669 A1 [0023]

Cited non-patent literature

• AJ Berkhout, D. de Vries, and P. Vogel. Acoustic control by wave field synthesis. Journal of the Acoustical Society of America, Vol. 93 (5): 2764-2778, May 1993 [0002]
• J. Daniel, Representation of Acoustic acoustics, application on the transmission and reproduction of sound and sound systems in the context of multimedia, PhD thesis, Université Paris 6, 2001 [0002]
• E. Hänsler, G. Schmidt, Topics in acoustic echo and noise control: selected methods for the cancellation of acoustical echoes, the reduction of background noise and speech processing, Springer-Verlag, Berlin 2006 [0003]
• H. Buchner, R. Aichner, and W. Kellermann, "TRINICON-based Blind System Identification with Application to Multiple-Source Localization and Separation," in Blind Speech Separation, S. Makino, H. Sawada, and T.-W. Lee, ed. Springer Netherlands, 2007, pp. 101-147 [0019]
• E. Hänsler, G. Schmidt, Topics in acoustic echo and noise control: selected methods for the cancellation of acoustical echoes, the reduction of background noise and speech processing, Springer-Verlag, Berlin 2006 [0023]
• H. Buchner, J. Benesty, and W. Kellermann, Generalized multichannel frequency-domain adaptive filtering: efficient realization and application to hands-free speech communication, Signal Processing, Vol. 85, No. 3, pp. 549-570, 2005 [0023]
• C. Faller and C. Tournery, Stereo acoustic echo control using a simplified echo path model, in Proc. IWAENC, 2006 [0023]
• H. Buchner, R. Aichner, and W. Kellermann, "TRINICON-based Blind System Identification with Application to Multiple-Source Localization and Separation," in Blind Speech Separation, S. Makino, H. Sawada, and T.-W. Lee, ed. Springer Netherlands, 2007, pp. 101-147. [0024]
• D. Ward, Microphone arrays: signal processing techniques and applications, Birkhäuser 2001 [0024]
• H. Buchner and S. Spors, A General Derivation of Wave-Domain Adaptive Filtering and Application to Acoustic Echo Cancellation, Proc. Asilomar Conference at Signals, Systems, and Computers, Pacific Grove, CA, USA, Oct. 2008 [0026]
• K. Helwani, H. Buchner, and S. Spors, Source-Domain Adaptive Filtering for MIMO Systems with Application to Acoustic Echo Cancellation, in IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2010. [0026]
• J. Dean and S. Ghemawat, "MapReduce: simplified data processing at large clusters", 6th Symposium on Operating Systems Design and Implementation, pp. 107-113, 2004 [0026]

Claims (15)

A method of reproducing a sound field emanating from a virtual source using a plurality of mobile terminals each having at least one electroacoustic transducer in a local space, comprising the steps Connecting the terminals through a data network, Determining the respective positions of the terminals in the local space, determining driving functions for the electroacoustic transducers for reproducing the sound field on the basis of a signal transmitted over the data network in dependence on the position of the virtual source and the determined positions of the terminals in the local space and Playing the sound field through the electroacoustic transducers in the terminals. The method of claim 1, wherein the terminals are interconnected by a star-shaped network with an external or local server or by a fully meshed network. The method of claim 1 or 2, wherein the terminals are connected by a wireless network, in particular a mobile network such as GSM, GPRS, UMTS or LTE, WLAN or Bluetooth. Method according to one of the preceding claims, wherein a location method is used for determining the position of the terminals, for example a video-based location method, an audio-based location method and / or a location method using GPS. Method according to one of the preceding claims, wherein the position of the terminals is determined by manual input by the user. The method of any one of the preceding claims, further comprising the step of: Synchronizing the playback by the individual terminals to compensate for the delay caused by the transmission of the signal. The method of any preceding claim, wherein the method further comprises the step of: Selecting the content of the sound field to be played by one of the terminals. The method of claim 7, wherein the signal transmitted over the data network represents the content of the sound field to be reproduced. Method according to claim 7 or 8, wherein the content of the sound field to be reproduced is the content of an audio signal from a distant room, for example the audio signal during a telephone call, for example by means of GSM or VoIP. A method according to claim 7, wherein the content of the sound field to be reproduced is an audio file stored in all the terminals. The method of claim 10, wherein the signal sent over the data network is a time pointer to the audio file. A method for full-duplex communication between a remote room and a local room using a plurality of mobile terminals each having an electro-acoustic transducer in the local space, with the steps Reproducing a sound field emanating from a virtual source with the method according to claim 9 and Picking up the acoustic signal to be transmitted from the local space to the distant space by the electroacoustic transducers of the terminals. The method of claim 12, wherein the local, the external server, and / or the individual mobile terminals comprise an echo cancellation device. Apparatus for carrying out a method according to any one of the preceding claims, wherein the apparatus comprises a plurality of mobile terminals, each having an electro-acoustic transducer, which are connected via a data network. The device of claim 14, wherein the mobile terminals are smartphones.

Images