Search

EP-4032320-B1 - VOLUME DEPENDENT AUDIO COMPENSATION

EP4032320B1EP 4032320 B1EP4032320 B1EP 4032320B1EP-4032320-B1

Inventors

  • PHILIPSSON, JOHN
  • MARTINSON, Roger

Dates

Publication Date
20260506
Application Date
20201112

Claims (15)

  1. A method (100) for real-time processing of an audio stream (33) in an audio speaker arrangement (20) comprising at least one audio speaker (21), the method (100) comprising: receiving (110) an audio-stream (33) and a playback volume (15) associated with the audio-stream (33), determining (120) a playback sound pressure level (52) of the audio speaker (21) based on the playback volume (15) and pre-stored acoustic transfer data (50), wherein the pre-stored acoustic transfer data (50) is associating the playback volume (15) with the playback sound pressure level (52) of the audio speaker (21), generating (130) a compensation filter (320) based on amplitude compensation data (330) associated with the determined playback sound pressure level (52), and filtering (140) the audio-stream (33) using the compensation filter (320) and applying a gain factor to the audio stream (33) generating a compensated audio-stream (27), wherein the gain factor is configured such that a total audio power accumulated across all frequencies of the audio-stream (33) is preserved regardless of the compensation filter (320, 320'), wherein the gain factor is generated based on an estimated signal frequency content of an average audio-stream and an estimated audio power of the average audiostream at a master level (12), wherein the master level (12) is the level of an initial recording of an audio-stream, wherein the average audio-stream and its frequency distribution is calculated by analyzing audio-streams, and providing (150) the compensated audio-stream (27) to the audio speaker (21).
  2. The method (100) according to claim 1, wherein the amplitude compensation data (330) is based on equal loudness contours (300).
  3. The method (100) according to any of the preceding claims, wherein the amplitude compensation data (330) comprises amplitude compensation values (335) for a number of different frequencies and a number of different playback volumes (15).
  4. The method (100) according to any of the preceding claims, further comprising the steps of: receiving (110) an updated playback volume (15') that is different from the playback volume (15), determining (120') an updated playback sound pressure level (52') of the audio speaker (21) based on the updated playback volume (15') and the pre-stored acoustic transfer data (50), generating (130') an updated compensation filter (320) based on amplitude compensation data (330) associated with the determined updated determined playback sound pressure level (52'), filtering (140) the audio-stream (33) to generate the compensated audio-stream (27) using the updated compensation filter (320), and providing (150) the compensated audio-stream (27) using the updated compensation filter (320) synchronized with the updated playback volume (15') to the audio speaker (21).
  5. The method (100) according to any of the preceding claims, wherein the audio-stream (33) is filtered in audio frames (23).
  6. The method (100) according to claim 5 when depending on claim 4, wherein the steps of determining (120') an updated playback sound pressure level (52) and generating (130') an updated compensation filter (320) are executed during an intermediate time (T I ) between the filtering of two consecutive audio frames (23).
  7. The method (100) according to any of the preceding claims, wherein the step of generating (130, 130') a compensation filter (320) is additionally based on a hearing profile (340) associated with a user (40) of the audio speaker arrangement (20).
  8. The method (100) according to any of the preceding claims, wherein the audio-stream (33) is formed in a channelized audio format.
  9. The method (100) according to claim 8, wherein the channelized audio format comprises a left channel audio-stream (33L) and a right channel audio-stream (33R), and wherein the step(s) of generating (130, 130') the compensation filter (320) further comprises generating (130, 130') a left compensation filter (320) for the left channel audio-stream (33L) and a right compensation filter (320) for the right channel audio-stream (33R).
  10. The method (100) according to any of the preceding claims, wherein the compensation filter (320) is a digital filter with a linear phase response, such as a FIR filter.
  11. An audio speaker arrangement (20) comprising at least one audio speaker (21), at least one memory (700), at least one external interface (30), and at least one controller (200) operatively connected to said at least one audio speaker (21), said at least one memory (700) and said at least one external interface (30); wherein said at least one controller (200) being configured to, in real-time: receive (110), via the external interface (30), an audio-stream (33) and a playback volume (15) associated with the audio-stream (33), determine (120) a playback sound pressure level (52) of the audio speaker (21) based on the playback volume (15) and pre-stored acoustic transfer data (50), wherein the pre-stored acoustic transfer data (50) is associating the playback volume (15) with the playback sound pressure level (52) of the audio speaker (21), generate (130) a compensation filter (320) based on amplitude compensation data (330) associated with the determined playback sound pressure level (52), filter (140) the audio-stream (33) using the compensation filter (320) and applying a gain factor to the audio stream (33) generating a compensated audio-stream (27), wherein the gain factor is configured such that a total audio power accumulated across all frequencies of the audio-stream (33) is preserved regardless of the compensation filter (320, 320'), wherein the gain factor is generated based on an estimated signal frequency content of an average audio-stream and an estimated audio power of the average audiostream at a master level (12), wherein the master level (12) is the level of an initial recording of an audio-stream, wherein the average audio-stream and its frequency distribution is calculated by analyzing audio-streams, and provide (150) the compensated audio-stream (27) to the audio speaker (21).
  12. The audio speaker arrangement (20) of claim 11, wherein the external interface (30) is a wireless interface.
  13. The audio speaker arrangement (20) of claim 11 or 12, wherein the controller (200) is further configured to perform the method (100) for real-time processing of the audio-stream (33) according to any of the claims 2 to 10.
  14. A computer program product (600) configured to, when executed by a controller (200), cause the controller (200) in real-time to: receive (110) an audio-stream (33) and a playback volume (15) associated with the audio-stream (33), determine (120) a playback sound pressure level (52) of an audio speaker (21) based on the playback volume (15) and pre-stored acoustic transfer data (50), wherein the pre-stored acoustic transfer data (50) is associating the playback volume (15) with the playback sound pressure level (52) of the audio speaker (21), generate (130) a compensation filter (320) based on amplitude compensation data (330) associated with the determined playback sound pressure level (52), and filter (140) the audio-stream (33) using the compensation filter (320) and applying a gain factor to the audio stream (33) generating a compensated audio-stream (27), wherein the gain factor is configured such that a total audio power accumulated across all frequencies of the audio-stream (33) is preserved regardless of the compensation filter (320, 320'), wherein the gain factor is generated based on an estimated signal frequency content of an average audio-stream and an estimated audio power of the average audiostream at a master level (12), wherein the master level (12) is the level of an initial recording of an audio-stream, wherein the average audio-stream and its frequency distribution is calculated by analyzing audio-streams, and provide (150) the compensated audio-stream (27) to the audio speaker (21).
  15. The computer program product (600) of claim 14, further configured to cause the controller (200) to perform the method (100) for real-time processing of the audio-stream (33) according to any of the claims 2 to 10.

Description

TECHNICAL FIELD The present invention relates to compensation of audio and more precisely to a method and a device for compensating an audio-stream in an audio speaker arrangement. BACKGROUND Since the introduction of portable electronic equipment such as the Walkman™ in the late 1970's, the availability and quality of portable audio has been ever increasing. With the smartphone, every user of such a device has access to an almost endless supply of music and other audio content. The audio content will typically be experienced through a pair of headphones or other portable speakers. The audio content is intended to sound in a specific way, e.g. the author and the producer of the content have a certain idea of how the content should be perceived by a listener. The same can be said with regards to the speakers and the headphone. The designers of those products typically have a vision of how their product should alter the audio content in order to create e.g. a brand specific sound or feeling. One solution to this problem is to personalize a listening experience by having a listener perform a hearing test and adjust the audio to compensate for the result of the hearing test. Implementations of this concept is presented in US2006215844 wherein an audio signal is filtered depending on a hearing profile of the listener. The prior art is focused on the threshold for hearing, i.e. the lowest sound pressure audible to the listener. However, the audio played is rarely played at a volume that corresponds to the threshold for hearing. A prior art real-time processing of an audio stream is disclosed in US 2017/133 992 A1. In view of the above, it is evident that there is room for improvements. SUMMARY An object of the present invention is to provide a new type of audio compensation which is improved over prior art and which eliminates or at least mitigates the drawbacks discussed above. More specifically, an object of the invention is to provide an audio compensation that is determined without prior knowledge of the audio signal and is applicable to streamed audio. These objects are achieved by the technique set forth in the appended independent claims with preferred embodiments defined in the dependent claims related thereto. According to a first aspect of the invention, a method for real-time processing of an audio-stream in an audio speaker arrangement is presented. The audio speaker arrangement comprises at least one audio speaker. The method comprises receiving an audio-stream and a playback volume associated with the audio-stream. From this, a playback sound pressure level of the audio speaker is determined based on the playback volume and pre-stored acoustic transfer data. The pre-stored acoustic transfer data is associating the playback volume with the playback sound pressure level of the audio speaker. The method further generates a compensation filter based on amplitude compensation data associated with the determined playback sound pressure level and filters the audio-stream using the compensation filter and applies a gain factor to the audio stream generating a compensated audio-stream. The gain factor is configured such that a total audio power of the audio stream is preserved regardless of the compensation filter. The compensated audio-stream is provided to the audio speaker. In one embodiment of the method, the amplitude compensation data is based on equal loudness contours. By basing the amplitude compensation data on equal loudness, the audio-stream will be perceived in the same way regardless of playback sound pressure level. In another embodiment, the amplitude compensation data comprises amplitude compensation values for a number of different frequencies and a number of playback volumes. By adding frequency dependencies, the audio-stream will be perceived in the same way regardless of playback sound pressure level. In an additional embodiment, the method further comprises the steps of receiving an updated playback volume, that is different from the playback volume, and determining an updated playback sound pressure level of the audio speaker based on the updated playback volume and the pre-stored acoustic transfer data. An updated compensation filter is generated based on amplitude compensation data associated with the determined updated playback sound pressure level. The audio-stream is filtered to generate the compensated audio-stream using the updated compensation filter. The compensated audio-stream is provided synchronized with the updated playback volume to the audio speaker using the updated compensation filter. By enabling the audiostream to be provided synchronized with the change in playback volume will allow for the volume compensated audio-stream to be delivered to a user without disruption when the playback volume is changed. In one embodiment, the step of filtering further comprises applying a gain factor to the audio stream generating a compensated audio-stream. The gain factor is confi