Method for calibrating a portable audio device, calibration system for a portable audio device, and associated computer program product

Abstract

Method for calibrating a portable audio device, calibration system for a portable audio device and associated computer program product. The present invention relates to a method for calibrating a portable audio device 10 comprising at least one portable microphone 20, 25 and a portable loudspeaker 15, the portable audio device 10 being arranged in a housing 95 comprising a calibration system 75. The calibration system comprises a calibration microphone 80, a calibration loudspeaker 85, and an electronic calibration device 90.The method comprises: - a phase of transmitting an audio signal to the portable speaker or the calibration speaker for its transmission, - a calibration phase comprising: - receiving a signal resulting from the acquisition of the audio signal emitted by the portable microphone or the calibration microphone, - determining a calibration filter H_calib_SPK; H_calib_FB; H_calib_FF, from the received signal, - a phase of implementing the determined calibration filter in the portable audio device. Figure for the abbreviation: [Figure 2].

Description

Title of the invention: Method for calibrating a portable audio device, system for calibrating a portable audio device and associated computer program product

[0001] The present invention relates to a method for calibrating a portable audio device.

[0002] The invention also relates to a computer program product and a system for calibrating a portable audio system.

[0003] The invention relates to the field of manufacturing portable audio devices.

[0004] The term "portable audio device" means a device configured to play, to a user of the device, audio content originating from an audio source, the device being transportable with the audio source. The audio source is for example an MP3 player or a smartphone. The portable audio device is an in-ear device, also called a pair of earphones. Alternatively, the portable audio device is for example an extra-auricular device, also called an audio headset.

[0005] In a manner known per se, a portable audio device comprises at least one portable speaker and preferably for each portable speaker, an internal portable microphone and at least one external portable microphone.

[0006] The portable speaker is configured to reproduce sound content from the source.

[0007] In a manner known per se, the internal portable microphone is oriented towards the portable loudspeaker, and intended to be positioned between said portable loudspeaker and an ear of the user. The portable microphone is configured to acquire the audio signal broadcast by the portable loudspeaker.

[0008] The external nomadic microphone is disposed on a periphery of the nomadic audio device and oriented towards an exterior of the device. The external nomadic microphone is configured to acquire the audio stream surrounding the nomadic audio device, such as the user's speech.

[0009] In a manner known per se, the portable loudspeaker, the external portable microphone and the portable loudspeaker, together with the source, allow a hands-free kit telephony functionality to be achieved. In such a functionality, the user's words are acquired by the external portable microphone and the words of a third party with whom the user is communicating are reproduced by the portable loudspeaker.

[0010] Furthermore, in a manner known per se, the portable loudspeaker, the internal portable microphone and the external portable microphone make it possible to implement a func- tionality active reduction of surrounding noise the portable audio device.

[0011] According to this functionality, the surrounding noise is acquired by the external nomadic microphone. This noise is processed by a filter and a signal is transmitted to the nomadic loudspeaker to cancel out the surrounding noise having passed through the audio device. In addition, the internal nomadic microphone makes it possible to control, by a respective filter, that a residual noise at the level of an eardrum of the user is minimal. Feedback is performed on the nomadic loudspeaker depending on the signal acquired by the internal nomadic microphone.

[0012] It is therefore clear that the internal and external microphones and the loudspeaker are connected. Transfer functions between these elements can therefore be calculated.

[0013] In a manner known per se, each of the portable loudspeakers and the portable microphones has its own frequency response.

[0014] Loudspeakers and microphones are generally mass-produced components. It is therefore known to determine, for each model, a respective nominal frequency response and a maximum dispersion between the frequency response of a specific component and the nominal response associated with the model of the component.

[0015] Conventionally, the maximum dispersion is for example equal to plus or minus 3dB.

[0016] In an industrial context, it is often necessary to design each filter of the portable audio device according to the nominal frequency response of each component without taking into account the dispersion between the behavior of the components actually in the audio headset and their nominal behavior. In addition, the design of each filter generally does not take into account defects during the assembly of the components to form the portable audio device.

[0017] However, when the frequency response of a component in the audio device varies significantly from the nominal frequency response or when the device has an assembly defect, the filters are no longer necessarily adapted to the device obtained.

[0018] The user experience is then at least degraded. In the worst scenarios, these defects are likely to lead to instability of the closed-loop transfer functions and therefore to the generation of feedback, which is particularly unpleasant for the user.

[0019] The present invention provides a method for calibrating a portable audio device, a computer program product and an associated calibration system making it possible to avoid such problems.

[0020] To this end, the invention relates to a method for calibrating a portable audio device, such as a headset or a pair of earphones, comprising at least one portable microphone and one portable speaker, the portable audio device being arranged in a box including a calibration system,

[0021] the calibration system comprising a calibration microphone, a calibration loudspeaker, and an electronic calibration device suitable for being connected to the calibration microphone and to the calibration loudspeaker, the method being implemented by the electronic calibration device and comprising:

[0022] - at least one transmission phase during which an audio signal is sent to one between the portable speaker and the calibration speaker, the audio signal being emitted by said speaker,

[0023] - a calibration phase during which at least one of the at least one mi portable microphone and the portable speaker is calibrated, the calibration phase including:

[0024] - a reception, from one of the at least one nomadic microphone and the mi calibration microphone, of a signal resulting from the acquisition by said microphone of the emitted audio signal,

[0025] - a determination of a calibration filter, from the signal received during the reception, the calibration filter being a digital filter comprising a gain whose value varies with frequency,

[0026] the calibration filter being:

[0027] - either specific to at least one portable microphone and configured to filter any signal acquired by said portable microphone,

[0028] - either specific to the portable loudspeaker and configured to filter any signal intended to be emitted by said portable speaker,

[0029] - an implementation phase, in the portable audio device, of the calibration filter determined during the calibration phase.

[0030] With the determination of a calibration filter and its implementation in the portable audio device, the performance of said device is improved based on the performance of the portable speaker and the portable microphone actually present in the portable audio device and not only from the nominal performance of the model of these components.

[0031] According to particular embodiments of the invention, the calibration method according to the invention comprises one or more of the following characteristics, taken in isolation or in all technically possible combinations:

[0032] - the transmission phase comprises a transmission step by the portable loudspeaker during from which an audio signal is sent to the portable speaker for transmission by said portable speaker,

[0033] the calibration phase comprising a step of calibrating the portable loudspeaker,

[0034] upon receipt of said calibration step, the audio signal is received from the calibration microphone,

[0035] when determining said calibration step, the determined calibration filter is specific to the portable loudspeaker, and called the loudspeaker calibration filter,

[0036] - the portable audio device comprises two portable microphones, the two mi portable microphones comprising an internal portable microphone facing the portable speaker, and an external portable microphone facing outside the portable audio device,

[0037] - the calibration phase further comprises a microphone calibration step internal nomad,

[0038] upon receiving the internal mobile microphone calibration step, the audio signal is received from the internal mobile microphone,

[0039] when determining the calibration step of the internal mobile microphone, the determined calibration filter is specific to the internal mobile microphone, and called the calibration filter of the internal microphone,

[0040] - the step of calibrating the internal portable microphone further comprises:

[0041] - an application of the loudspeaker calibration filter to the signal received during the reception of the calibration step of the internal nomadic microphone, to form a filtered signal,

[0042] during the step of determining the calibration phase of the internal nomadic microphone, the calibration filter of the internal microphone is determined as a function of the filtered signal,

[0043] - the transmission phase comprises a step of transmission by the loudspeaker of calibration, during which an audio signal is sent to the calibration speaker for transmission by said calibration speaker,

[0044] the calibration phase comprising a step of calibrating the external portable microphone,

[0045] upon receiving the calibration step of the external mobile microphone, the audio signal is received from the external mobile microphone,

[0046] when determining the calibration step of the external mobile microphone, the determined calibration filter is specific to the external mobile microphone, and called the calibration filter of the external microphone,

[0047] - the portable audio device further comprises a control unit comprising a summing device, a first connection connecting the summing device to the portable loudspeaker, a second connection connecting the internal portable microphone to the summing device, and a third connection connecting the external portable microphone to the summing device,

[0048] the portable audio device further comprising a first predetermined filter on the second connection, a second predetermined filter on the third connection,

[0049] during the implementation phase, the or at least one of the calibration filters being implemented on one of the first, second and third connections,

[0050] - the or each determination sub-step comprises:

[0051] - a calculation of a frequency error between the corresponding received signal and a signal predetermined reference,

[0052] - a detection of extremum(s) of the frequency error,

[0053] - for at least one detected extremum:

[0054] - a calculation of a frequency band around the extremum,

[0055] - a determination of a unitary filter that can be configured as a function of the error fre quantially in the frequency band,

[0056] - a generation of the calibration filter from the or each para unitary filter meterable,

[0057] each configurable unitary filter preferably comprising the following adjustable parameters: a central frequency, a maximum gain, and a quality factor,

[0058] preferably, each configurable unitary filter further comprising a type of filter chosen from a group consisting of: a peak or trough filter, a stepped low-pass filter, and a stepped high-pass filter,

[0059] - for each extremum detected, when calculating the frequency band, each frequency band is delimited by a lower limit and an upper limit,

[0060] the lower limit of the frequency band being the upper limit of the previous frequency band,

[0061] the upper limit of the frequency band being calculated according to the following equation:

[0062] = k-MO )

[0063] where

[0064] - f1 is the frequency of the extremum detected in said frequency band B;, ■' ext

[0065] - fl+l is the frequency of the extremum detected in said frequency band ■' ext next,

[0066] - |&r(4J a~

[0067] - ^ErrÇf* is the norm of the error evaluated at the frequency of the detected extremum in said frequency band, and

[0068] - is the norm of the error evaluated at the frequency of the detected extremum in the following frequency band; and

[0069] - the portable audio device comprises at least two audio channels, each channel audio comprising at least one respective portable microphone and one respective portable speaker,

[0070] the steps of the calibration method being iterated for each audio channel.

[0071] The present invention also relates to a computer program product comprising software instructions, which, when executed by a computer, implement the calibration method as described above.

[0072] The present invention also relates to a system for calibrating a portable audio device, such as a headset or a pair of earphones, the portable audio device comprising at least one portable microphone, a portable loudspeaker, the portable audio device being suitable for being placed in the calibration system,

[0073] the calibration system comprising a calibration microphone, a calibration speaker, and an electronic calibration device capable of executing software instructions of a computer program product described above.

[0074] The invention will be better understood and other advantages thereof will appear more clearly on reading the description which follows, given solely by way of non-limiting example, and made with reference to the appended drawings in which:

[0075] - [Fig.l] [Fig.l] is a schematic view of a portable audio device before the application of a calibration method according to the invention;

[0076] - [Fig.2] [Fig.2] is a schematic view of the portable audio device of [Fig.l] when implementing the calibration method according to the invention; and

[0077] - [Fig.3] [Fig.3] is a flowchart of the calibration method according to the invention.

[0078] In [Fig.l], a portable audio device 10 is shown.

[0079] In the example of [Fig.l], the portable audio device 10 is a pair of earphones. In a variant not shown, the portable audio device is an audio headset.

[0080] The portable audio device 10 optionally comprises a first audio channel CH1 and a second channel CH2.

[0081] For example, the first audio channel CH1 is the left channel, i.e. intended to broadcast content to a user's left ear. In this example, the second channel CH2 is a right channel, i.e. intended to broadcast content to a user's right ear.

[0082] It is clear that an inverted configuration is also possible.

[0083] The portable audio device 10 preferably comprises, for each audio chain CH1, CH2, a portable loudspeaker 15 and at least one portable microphone 20, 25.

[0084] Furthermore, the portable audio device 10 comprises, preferably for each audio chain CH1, CH2, a connection point 30 with an audio source not shown, and a control unit 33.

[0085] The portable speaker 15 is configured to broadcast content originating from the audio source. Optionally, the portable speaker 15 is configured to broadcast content determined by the control unit 33, for example in order to perform an active noise reduction functionality.

[0086] The portable microphone 20, 25 is configured to acquire an audio signal. Preferably, the portable audio device 10 comprises, advantageously for each channel CH1, CH2, an internal portable microphone 20 and at least one external portable microphone 25.

[0087] The internal nomadic microphone 20, also called internal microphone 20, is configured to acquire the audio signal between the nomadic speaker 15 and the user's ear, in particular the audio signal broadcast by the nomadic speaker 15. It is then understood that, when the audio device is worn by a user, the internal nomadic microphone 20 is arranged between the nomadic speaker 15 and an eardrum of the user. The internal nomadic microphone 20 is then configured to acquire the audio signal heard by the user.

[0088] The external nomadic microphone 25, also called external microphone 25, is configured to acquire an audio stream surrounding the nomadic audio device 10. For this purpose, the external microphone 25 is arranged on a periphery of the nomadic audio device 10 and oriented towards an exterior of said device 10.

[0089] The connection point 30 is configured to receive, from the audio source, content to be broadcast by the portable speaker 15. The connection point 30 is for example configured to communicate with the audio source by a wired connection or, as shown in [Fig.l], by a wireless connection following a predefined protocol, such as the Bluetooth or WIFI protocol.

[0090] The control unit 33 is configured to control the signal sent to the portable loudspeaker 15 as a function of the signal received from the connection point 30, the signal acquired by the internal microphone 20 and the signal acquired by the external microphone 25.

[0091] The control unit 33 is connected to the portable loudspeaker 15, to the internal microphone 20, to the external microphone 25 and to the connection point 30. The control unit 33 is for example an electronic card comprising a plurality of software modules.

[0092] The control unit 33 comprises a summer 35, a first connection 50 connecting the summer 35 and the portable loudspeaker 15, a second connection 55 connecting the internal microphone 20 and the summer 35, a third connection 60 connecting the external microphone 25 and the summer 35 and preferably a fourth connection 65 connecting the connection point 30 and the summer 35.

[0093] The control unit 33 comprises for example a conversion and amplification block 70 on the first connection 50. The block 70 is for example formed of a DAC (Digital-to-Analog Converter) converter capable of converting a digital signal, coming from the adder 35 into an analog signal for exciting the portable loudspeaker 15, and of an amplifier of the analog signal.

[0094] The control unit 33 preferably further comprises a first predetermined filter HFB on the second connection 55. The first predetermined filter HFB is configured to filter the signal acquired from the internal nomadic microphone 20 before sending it to the adder 35.

[0095] The control unit 33 preferably further comprises a second predetermined filter HFF on the third connection 60. The second predetermined filter HFF is configured to filter the signal acquired from the external nomadic microphone 25 before transmitting it to the adder 35.

[0096] The first HFB and second HFF predetermined filters are for example designed as a function of a nominal frequency response of the internal 20 and external 25 microphones, as provided by a manufacturer of said microphones and measured on reference prototypes.

[0097] The control unit 33 further comprises an HEQ equalizer on the fourth connection 65. The HEQ equalizer is configured to perform an equalization of the signal coming from the connection point 30 before transmitting it to the summer 35.

[0098] In [Fig.2], the portable audio device 10 is coupled with a calibration system 75 for calibrating said audio device 10. In [Fig.2], only one chain CH1, CH2 of the portable audio device 10 is shown.

[0099] The calibration system 75 comprises a calibration microphone 80, a calibration loudspeaker 85 and an electronic calibration device 90 suitable for being connected to the calibration microphone 80 and to the calibration loudspeaker 85.

[0100] To carry out the calibration of the portable audio device 10, said device 10, the calibration microphone 80 and the calibration loudspeaker 85 are placed in an acoustically sealed box 95. The calibration microphone 85 is positioned contiguously with the internal portable microphone 20, for example within an artificial ear (not shown). And the calibration loudspeaker 85 is positioned outside the portable audio device 10.

[0101] The calibration microphone 80 and the calibration speaker 85 are connected to the electronic calibration device 90 as visible in [Fig.2] in solid line.

[0102] The portable loudspeaker 15, the internal microphone 20, and the external microphone 25 are connected to the electronic calibration device 90, as shown in [Fig.2] in phantom.

[0103] The electronic calibration device 90 is located outside the box 95. The electronic calibration device 90 is for example a computer comprising a memory 96 and a processor 97 capable of executing software instructions stored in the memory 96.

[0104] The memory 96 stores a computer program comprising software instructions which, when executed by the processor 97, implement a method of calibrating the portable audio device 10 as described below.

[0105] With reference to [Fig. 3], the method for calibrating the portable audio device 10 comprises a transmission phase during which an audio signal is sent to one of the portable loudspeakers 15 and the calibration loudspeaker 85.

[0106] The transmission phase preferably comprises first of all a first transmission step PE_nomad during which the calibration device 90 sends to the nomad loudspeaker 15, an audio signal, called first audio signal. Following this first transmission step PE, the nomad loudspeaker 15 transmits the first audio signal towards the internal nomad microphone 20 and the calibration microphone 80. The first audio signal is for example a continuous sinusoidal sweep. During the transmission of the first audio signal by the nomad loudspeaker 15, the calibration microphone 80 and the internal nomad microphone 20 acquire the transmitted signal.

[0107] Then, the calibration method comprises a phase of calibrating at least one of the portable loudspeaker 15, the internal portable microphone 20 and the external portable microphone 25.

[0108] The calibration phase preferably comprises a PC_SPK step for calibrating the portable loudspeaker 15.

[0109] The calibration step PC_SPK comprises a first reception sub-step 112, during which the electronic calibration device 90 receives, from the calibration microphone 80, the signal resulting from the acquisition by said microphone of the first audio signal emitted by the portable loudspeaker 15.

[0110] The calibration step PC_SPK further comprises a first determination sub-step 114, during which the calibration device 90 determines a first calibration filter H_calib_SPK from the signal received from the calibration microphone 80. The first calibration filter H_calib_SPK is a digital filter comprising a gain whose value evolves in frequency. “A gain whose value evolves in frequency” means that the gain takes distinct values ​​depending on the frequency of the signal which is supplied to the associated filter. The first calibration filter H_calib_SPK is specific to the portable loudspeaker 15.

[0111] For this, the first determination sub-step 114 advantageously comprises the calculation of a frequency error Err(f) between the signal received during the first reception sub-step 112 and a first predetermined reference signal.

[0112] The first reference signal corresponds to the signal that the calibration microphone 80 should have acquired if the frequency response of the portable loudspeaker 15 was exactly the nominal frequency response of a reference product.

[0113] To calculate this error Err(f), in the following example, the calibration device 90 calculates a Fourier transform of the signal received from the calibration microphone

[0114]

[0115]

[0116]

[0117]

[0118]

[0119]

[0120]

[0121]

[0122]

[0123]

[0124]

[0125]

[0126] 80. In a manner known per se, the signal resulting from the Fourier transform comprises an amplitude and a phase evolving respectively in frequency. Still when calculating the frequency error Err(f), the calibration device 90 subtracts for example the amplitude of the Fourier transform from the amplitude of the first predetermined reference signal, for example according to the following equation: [Math.l] Err(f) - dB(Target ( / )) - dB(Init(f)) where Targedf) is the first predetermined reference signal, is the Fourier transform of the received signal, and dBQ is the operation converting amplitude to decibels. Then, still during the determination sub-step 114, the calibration device 90 advantageously performs a detection of extremum(s) f' in the frequency error exi Err(f). For this purpose, the calibration device 90 calculates, for example, a derivative of the frequency error Err(f) as a function of the frequency and determines each frequency y* for which said derivative is zero. Still during the determination sub-step 114, the calibration device 90 preferentially calculates, for each determined extremum y', a frequency band B; around said extremum f'. Two successive frequency bands B, H ■' ext B; are contiguous. In other words, each frequency band B; includes a lower bound f{ and an upper bound yL Also, the lower bound f of the frequency band B; is the upper bound f' of the previous frequency band B, h The lower bound y * of the first frequency band Bi is the minimum frequency for which the frequency error Err(f) is defined. Then, the upper band f1 of the frequency band B; is calculated according to the following equation: [Math.2] p _ iQrT-'e.ïJ + '“h1 where \En\f ,)j has _ After calculating the frequency bands B;, the entire frequency domain for which the frequency error Err(f) is defined is divided into a plurality of frequency bands B;. Each frequency band B; contains an extremum f of the error fre-■' ext quential Err(f). Still during the determination sub-step 114, the calibration device 90 determines for example, for each extremum f1, a parameterizable unitary filter H; as a function of the frequency error Err(f) in the frequency band B; associated with said extremum. J ext

[0127] Each parameterizable unitary filter H; is for example a peak-notch filter known per se. This filter H; is characterized by three parameters: a central frequency f1, a gain A1 and a quality factor Q'. In a manner known per se, the central frequency f1 corresponds to the frequency for which the parameterizable unitary filter H; has the highest gain. The gain A' corresponds to the maximum gain of the parameterizable unitary filter, for example its gain at the central frequency f1 for a Peak-notch filter. The quality factor Ql quantifies the size of the bandwidth of the configurable unit filter.

[0128] Then, the calibration device 90 advantageously initializes the parameters of the parameterizable unitary filter H; and calculates for example a modified frequency error, in the band B; according to the following equation:

[0129] [Math.3] Err(f) = dB( Target ( f ) ) -dB^H,( f ) *lnit( f ))

[0130] where / is between f1 and fy^

[0131] The preceding equation corresponds to the frequency error if the received signal was filtered by the parameterizable unitary filter H;.

[0132] Then, the calibration device 90 optionally applies an optimization algorithm to the parameter of the configurable unitary filter in order to minimize an error in the least squares sense defined from the modified error and which is written for example in the following form:

[0133] [Math.4] | (dB( Target (f))- dB( ( f ) *Init ( / ) ) ) 2

[0134] where N is the number of samples of the signal received between frequencies f* and y'L

[0135] The optimization algorithm is for example a gradient descent algorithm.

[0136] Following the application of the optimization algorithm, the calibration device 90 obtains optimized parameters p , and the parameterizable unitary filter H;.

[0137] Advantageously, the calibration device 90 thus determines a unitary filter para metrical H; respective for each extremum f1, and therefore for each band of J ext frequency B; associated.

[0138] Then, the calibration device 90 generates the first calibration filter H_calib_SPK, for example from each parameterizable unitary filter H; determined previously, for example by putting in series each of the configurable unitary filters H;.

[0139] The calibration phase optionally further comprises a PC_FB calibration step of the internal portable microphone 20.

[0140] As indicated previously, following the emission of the first audio signal by the portable speaker 15, the internal portable microphone 20 acquires the first audio signal emitted.

[0141] The step PC_FB of calibrating the internal nomadic microphone 20 comprises a second reception sub-step 122 during which the calibration device 90 receives the first audio signal acquired by the internal microphone 20.

[0142] Preferably, the step PC_FB of calibrating the internal microphone 20 comprises a sub-step 123 of applying the first calibration filter H_calib_SPK to the signal received from the internal microphone 20, to form a filtered signal.

[0143] The step PC_calib_FB comprises a second determination sub-step 124, during which the calibration device 90 determines a second calibration filter H_calib_FB from the filtered signal. The second calibration filter H_calib_FB is also a digital filter comprising a gain whose value evolves frequently. The second calibration filter H_calib_FB is specific to the internal portable microphone 20.

[0144] The second determination sub-step 124 is analogous to the first determination sub-step 114 except that the determination is carried out from the signal received by the internal nomadic microphone 20 filtered by the function H_calib_SPK and not from the signal received from the calibration microphone 80.

[0145] At the end of the second determination sub-step 124, the second calibration filter H_calib_FB is determined.

[0146] Then, the transmission phase comprises a PE_calib step of sending a second audio stream to be broadcast by the calibration loudspeaker 85. Following this PE_calib sending step, the second audio signal is emitted by the calibration loudspeaker 85. During this transmission, the second audio signal emitted is acquired by the external nomadic microphone 25.

[0147] Analogously to the first audio signal, the second audio signal is, for example, a continuous sinusoidal sweep.

[0148] Then, the calibration phase comprises for example a step PC_FF for calibrating the external portable microphone 25.

[0149] The step PC_FF of calibrating the external mobile microphone 25 comprises a third reception sub-step 132 during which the calibration device 90 receives the signal acquired by the external mobile microphone 25 following the emission of the second audio signal by the calibration loudspeaker 85.

[0150] The step PC_FF of calibrating the external mobile microphone 25 further comprises a third determination sub-step 134 during which a third calibration filter H_calib_FF is determined from the signal received from the external mobile microphone 25. The third calibration filter H_calib_FF is also a digital filter comprising a gain whose value evolves in frequency. The third calibration filter H_calib_FF is specific to the external mobile microphone 25.

[0151] The third determination sub-step 134 is for example analogous to the first determination sub-step 114 and to the second determination sub-step 124, except that the third calibration filter H_calib_FF is determined from the signal received from the external nomadic microphone 25 rather than from the signal received from the calibration microphone 80 or from the internal nomadic microphone 20.

[0152] At the end of the determination sub-step 134, the third calibration filter H_calib_FF is determined.

[0153] The calibration method further comprises an implementation phase PI preferably comprising a first implementation step 142 during which the first calibration filter H_calib_SPK is implemented in the portable audio device 10, more precisely on the first connection 50 of the control unit 33. For example, the first calibration filter H_calib_SPK is implemented between the adder 35 and the block 70.

[0154] The implementation phase PI further preferably comprises a second implementation step 144 during which the second calibration filter H_calib_FB is implemented in the portable audio device 10, and more particularly on the second connection 55 of the control unit 33. For example, the second calibration filter H_calib_FB is implemented between the internal portable microphone 20 and the first predetermined filter HFB.

[0155] The implementation phase PI further preferably comprises a third implementation step during which the third calibration filter H_calib_FF is implemented in the portable audio device 10, and preferably on the third connection 60 of the control unit 33. For example, the third calibration filter H_calib_FF is implemented between the external portable microphone 25 and the second predetermined filter HFF.

[0156] At the end of the calibration method according to the invention, the portable loudspeaker 15, the internal portable microphone 20, and an external portable microphone 25 are calibrated.

[0157] At the end of the PI implementation phase, values ​​of the or each calibration filter H_calib_SPK, H_calib_FB, H_calib_FF are stored in the control unit 33, and preferably in the memory 96 of the control unit.

[0158] When the portable audio device 10 comprises several audio chains CH1, CH2, the calibration method described above is iterated for each audio chain CH1, CH2.

[0159] According to a first variant, the step PE_calib of sending the second audio signal and the step PC_FF of calibrating the external nomadic microphone 25 are implemented prior to the step PE_nomade of sending the first audio signal.

[0160] According to a second variant not shown, the audio device 10, preferably each audio chain CH1, CH2, comprises a plurality of external nomadic microphones 25;. According to this second variant, the control unit 33 comprises a plurality of third connections 60; respectively connecting an internal nomadic microphone 25; and the adder 35. The control unit 33 comprises, on each of these connections 60;, a second respective predetermined filter HFF;.

[0161] According to this second variant, following the PE_calib step of sending the second audio signal to the calibration loudspeaker 85, each external nomadic microphone 25; acquires the second audio signal from said transmission. The calibration phase then comprises, for each external nomadic microphone 25, a PC_FF; step of calibrating said external nomadic microphone 25 during which a third filter H_calib_FFi respective to the external nomadic microphone 25; is determined.

[0162] According to the second variant, the implementation phase comprises, for each external nomadic microphone 25;, a third respective implementation step 146;, during which the third respective calibration filter H_calibr_FF; is implemented on the connection 60; associated with said external nomadic microphone 25;.

[0163] According to a third variant, for example combinable with one of the other variants, each parameterizable unitary filter H; is a stepped low-pass filter (from the English Lowshelf) or a stepped high-pass filter (from the English Highshelf). The parameters of the parameterizable unitary filter H; are the same as when it is of the Peak-notch type.

[0164] In a manner known per se, a Lowshelf or Highshelf filter defines a slope that is respectively decreasing or increasing. Each parameterizable unitary filter H; of the form Lowshelf or Highshelf is parameterized by a central frequency, a gain A' and a quality factor Q'. In a manner known per se, the central frequency corresponds to the frequency in the middle of the slope of the Lowshelf or Highshelf filter. The gain A1 corresponds to the maximum gain of the parameterizable unitary filter. The quality factor Q quantifies the steepness of the slope of the Lowshelf or Highshelf type filter.

[0165] According to a fourth variant, during each determination sub-step 114, 124, 134, a frequency pre-splitting is carried out in a predefined manner forming a plurality of frequency portions. The width of a frequency portion being greater than the width of a frequency band B;. A maximum number of parameterizable unit filters H; per frequency portion is predefined. When determining the frequency bands B;, bands B; are determined until the number of bands B; in a frequency portion reaches the maximum number of configurable unitary filters H; of said portion.

[0166] According to a fifth variant, only one of the portable speaker 15, the internal microphone 20 and the external microphone 25 is calibrated. In this case, the transmission phase only comprises one transmission step PE_nomade, PE_calib. In this case, the calibration phase only comprises one calibration step PC_SPK, PC_FB, PC_FF. Thus, a single calibration filter H_calib_SPK, H_calib_FB, H_calib_FF is determined and implemented in the portable audio device 10.

[0167] It is clear that the method could also comprise one or two calibrations among the three calibrations presented above.

[0168] The method according to the invention therefore makes it possible to avoid problems linked to the variance of the frequency responses of the components of the audio device compared to a nominal frequency response. Indeed, after application of the method, the behavior of each calibrated component is close to the nominal frequency response provided by the manufacturer for the model of said component.

[0169] Furthermore, the determination of calibration filter(s) specific to each of the portable loudspeaker 15, the internal portable microphone 20 and the external portable microphone 25 actually in the portable audio device 10 allows customization of the correction made and therefore further improves the user experience.

[0170] Furthermore, since the method according to the invention can be implemented by a computer program product, it does not require any human intervention. This method is thus rapid and efficient.

[0171] In a preferred embodiment, thanks to the application sub-step 123, the step PC_FB of calibrating the internal microphone 20 overcomes the error that the portable loudspeaker 15 had before its calibration. Indeed, the application step 123 has the same effect as if the signal acquired by the internal microphone 20 had been emitted by the calibrated portable loudspeaker 15. This step accelerates the calibration process since it makes it possible to carry out only a single emission by the portable loudspeaker 15 to calibrate both said portable loudspeaker 15 and the internal microphone 20.

Claims

Claims

1. A method for calibrating a portable audio device (10), such as a headset or a pair of earphones, comprising at least one portable microphone (20, 25) and a portable loudspeaker (15), the portable audio device (10) being arranged in a box (95) comprising a calibration system (75), the calibration system (75) comprising a calibration microphone (80), a calibration loudspeaker (85), and an electronic calibration device (90) suitable for being connected to the calibration microphone (80) and to the calibration loudspeaker (85), the method being implemented by the electronic calibration device (90) and comprising: - at least one transmission phase during which an audio signal is sent to one of the portable loudspeaker (15) and the calibration loudspeaker (85), the audio signal being emitted by said loudspeaker (15;85), - a calibration phase during which at least one of the at least one portable microphone (20, 25) and the portable loudspeaker (15) is calibrated, the calibration phase comprising:; - a reception (112; 122; 132), from one of the at least one nomadic microphone (20, 25) and the calibration microphone (80), of a signal resulting from the acquisition by said microphone (20, 25; 80) of the emitted audio signal, - a determination (114; 124; 134) of a calibration filter (H_calib_SPK; H_calilb_FB; H_calib_FF), from the signal received during the reception (112; 122; 132), the calibration filter (H_calib_SPK; H_calilb_FB; H_calib_FF) being a digital filter comprising a gain whose value varies in frequency, the calibration filter (H_calib_SPK; H_calilb_FB;H_calib_FF) being: - either specific to the at least one portable microphone (20, 25) and configured to filter any signal acquired by said portable microphone (20, 25), - or specific to the portable loudspeaker (15) and configured to filter any signal intended to be emitted by said portable loudspeaker (15), - an implementation phase (PI), in the portable audio device (10), of the calibration filter (H_calib_SPK; H_calilb_FB; H_calib_FF) determined during the calibration phase.;

2. Method according to claim 1, in which the transmission phase comprises a transmission step (PE_nomade) by the nomadic loudspeaker (15) during which an audio signal is sent to the loudspeaker nomadic speaker (15) for its emission by said nomadic speaker (15), the calibration phase comprising a step (PC_SPK) of calibrating the nomadic speaker (15), upon reception (112) of said calibration step (PC_SPK), the audio signal is received from the calibration microphone (80), upon determination (114) of said calibration step (PC_SPK), the determined calibration filter (H_calib_SPK) is specific to the nomadic speaker (15), and called speaker calibration filter (H_calib_SPK).

3. The method of claim 1 or 2, wherein the portable audio device (10) comprises two portable microphones (20, 25), the two portable microphones (20, 25) comprising an internal portable microphone (20) facing the portable speaker (30), and an external portable microphone (25) facing an exterior of the portable audio device (10).

4. Method according to claims 2 and 3, wherein the calibration phase further comprises a step (PC_FB) of calibrating the internal mobile microphone (20), upon reception (122) of the step (PC_FB) of calibrating the internal mobile microphone (20), the audio signal is received from the internal mobile microphone (20), upon determination (124) of the step (PC_FB) of calibrating the internal mobile microphone (20), the determined calibration filter (H_calib_FB) is specific to the internal mobile microphone (20), and is called the internal microphone calibration filter (H_calib_FB).

5. Method according to claim 4, in which the step (PC_FB) of calibrating the internal mobile microphone (20) further comprises: - an application (123) of the loudspeaker calibration filter (H_calib_SPK) to the signal received during the reception (122) of the step (PC_FB) of calibrating the internal mobile microphone (20), to form a filtered signal, during the step of determining (124) the calibration phase of the internal mobile microphone (PC_FB), the calibration filter of the internal microphone (H_calib_FB) is determined as a function of the filtered signal.

6. Method according to any one of claims 3 to 5, in which the transmission phase comprises a transmission step (PE_calib) by the calibration loudspeaker (85), during which an audio signal is sent to the calibration loudspeaker (85) for its transmission by said calibration loudspeaker (85), the calibration phase comprising a step (PC_FF) of calibrating the external mobile microphone (25), upon reception (132) of the step (PC_FF) of calibrating the external mobile microphone (25), the audio signal is received from the external mobile microphone (25), upon determination (134) of the step (PC_FF) of calibrating the external mobile microphone (25), the determined calibration filter (H_calib_FF) is specific to the external mobile microphone (25), and is called the external microphone calibration filter (H_calib_FF).

7. Method according to any one of claims 3 to 6, wherein the portable audio device (10) further comprises a control unit comprising a summer (35), a first connection (50) connecting the summer (35) to the portable loudspeaker (15), a second connection (55) connecting the internal portable microphone (20) to the summer (35), and a third connection (60) connecting the external portable microphone (25) to the summer (35), the portable audio device (10) further comprising a first predetermined filter (HFB) on the second connection (55), a second predetermined filter (HFF) on the third connection (60), during the implementation phase (PI), the or at least one of the calibration filters (H_calib_SPK; H_calib_FB; H_calib_FF) being implemented on one of the first (50), second (55) and third (60) connections.

8. Method according to any one of the preceding claims, in which the or each determination sub-step (114, 124, 134) comprises: - a calculation of a frequency error (Err(f)) between the corresponding received signal and a predetermined reference signal, - a detection of extremum(s) ( f' ) of the frequency error (Err(f)), J is - for at least one detected extremum ( ): ■ exî - a calculation of a frequency band (B;) around the extremum, - a determination of a parameterizable unitary filter (H;) as a function of the frequency error (Err(f)) in the frequency band (B;), - a generation of the calibration filter (H_calib_SPK; H_calib_FB; H_calib_FF) from the or each parameterizable unitary filter, each parameterizable unitary filter (H;) preferably comprising the following adjustable parameters: a central frequency (y'), a maximum gain (aO, and a quality factor (g'), preferably, each configurable unitary filter (H;) further comprising a type of filter chosen from a group consisting of: a peak or trough filter, a stepped low-pass filter, and a stepped high-pass filter.

9. Method according to claim 8, wherein for each detected extremum ( fl ), when calculating the frequency band (B;), each frequency band (B;) is delimited by a lower bound (y') and an upper bound (yp, the lower bound ( fl) of the frequency band (B;) being the upper bound (of the previous frequency band (Bj ।), the upper bound (y^) of the frequency band (B;) being calculated according to the following equation: yi _ where - f' is the frequency of the detected extremum in said frequency band (B;), - is the frequency of the detected extremum in said following frequency band (Bi+i), - ; IM / Jl ■ \Err^f' )| is the norm of the error evaluated at the frequency of the detected extremum in said frequency band (B;), and ■ |Err(fl+ *)| is the norm of the error evaluated at the frequency of the extremum detected in the following frequency band (Bi+i).

10. Method according to any one of the preceding claims, in which the portable audio device (10) comprises at least two audio chains (CH1, CH2), each audio chain (CH1, CH2) comprising at least one respective portable microphone (20, 25) and one respective portable loudspeaker (15), the steps of the calibration method being iterated for each audio chain (CH1, CH2).

11. A computer program product comprising software instructions, which, when executed by a computer, implement the phases of a calibration method according to any one of the preceding claims.

12. Calibration system (75) of a portable audio device (10), such as a headset or a pair of earphones, the portable audio device (10) comprising at least one portable microphone (20, 25), a portable speaker (15), the portable audio device (10) being suitable for being placed in the calibration system (75), the calibration system (75) comprising a calibration microphone (80), a calibration speaker (85), and an electronic calibration device (95) suitable for executing software instructions of a computer program product according to the preceding claim.

Images