In-vehicle audio adjustment method, in-vehicle infotainment system and readable storage medium

Abstract

The present invention provides a method for adjusting in-vehicle audio, an in-vehicle infotainment system and a readable storage medium, the method comprising: collecting in-vehicle audio signals, separating the in-vehicle audio signals, and outputting a plurality of separated signals; judging whether the plurality of separated signals contain a human voice signal, if the separated signal does not contain a human voice signal, inputting a noise signal and a first playback audio signal into a first side chain multi-band compressor, and outputting a gain playback audio signal; if the separated signal contains a human voice signal, inputting the human voice signal and the second playback audio signal into a second side chain multi-band compressor, and outputting a compressed playback signal. The in-vehicle infotainment system implements the in-vehicle audio adjustment method. The readable storage medium applies the in-vehicle audio adjustment method. Different multi-band compression methods are adopted according to different sounds, so that while maintaining the in-vehicle music atmosphere, the voices of people speaking in the car can be heard clearly or the noise does not conflict with the in-vehicle music.

Description

Technical Field

[0001] The present invention relates to the technical field of sound processing, and in particular to a method for adjusting vehicle audio, a vehicle infotainment system and a computer-readable storage medium. Background Art

[0002] At present, the in-vehicle infotainment systems on the market do not have the function of automatically adjusting the audio volume. Drivers usually need to adjust the volume manually, which is inconvenient for drivers to operate while driving. Some existing automatic adjustment technologies only adjust the overall volume change and cannot distinguish between speech in the car, music playing and external noise. When there is a conversation in the car, the loudness of the music may increase, making it difficult to hear the conversation in the car. When the external noise increases, the loudness of the music increases, making the music harsh and reducing the listening experience.

[0003] There is an existing method for controlling ambient sound in a smart cockpit of a car. The method collects ambient sound outside the car through the microphone array of the smart cockpit. When playing entertainment programs in the car, the volume of the entertainment programs is determined by the sound tester in the car, and the model is used to determine whether the volume of the entertainment programs masks the external ambient sound. If so, the volume is automatically adjusted down. When the entertainment programs played in the car are not suitable for reducing the volume, the external ambient sound is enhanced and played through the car audio system. When the volume of the entertainment programs played in the car is greater than the preset threshold and human voice input is detected, the algorithm is used to identify whether the human voice input is singing or speaking, and the speaking voice is enhanced and the singing voice is weakened. However, this ambient sound method only adjusts the volume of the entertainment programs, and does not adjust the frequency bands of the volume of the entertainment programs according to different sounds, thereby weakening the audience's hearing. Summary of the invention

[0004] The first object of the present invention is to provide a method for adjusting vehicle audio for performing multi-band adjustment of music according to different sounds.

[0005] A second object of the present invention is to provide an in-vehicle infotainment system that implements the in-vehicle audio adjustment method.

[0006] A third object of the present invention is to provide a readable storage medium to which the above-mentioned vehicle audio adjustment method is applied.

[0007] To achieve the first purpose of the present invention, the present invention provides a method for adjusting in-vehicle audio, which is characterized in that the method includes: collecting in-vehicle audio signals, separating the in-vehicle audio signals, and outputting multiple separated signals, wherein the multiple separated signals include noise signals and playback audio feedback signals; judging whether the multiple separated signals contain human voice signals, and if the separated signals do not contain human voice signals, inputting the noise signal and the first playback audio signal into a first side chain multi-band compressor, dividing the noise signal into noise frequency band signals of a first preset number of frequency bands, dividing the first playback audio signal into first playback audio frequency band signals of a first preset number of frequency bands, and judging whether the level values ​​of the first preset number of noise frequency band signals are is greater than a first preset value. If so, a gain is performed on the first playback audio frequency band corresponding to the noise frequency band signal whose level value is less than the first preset value, and a gained playback audio signal is output; if the separated signal is a vocal signal, the vocal signal and the second playback audio signal are input into a second side chain multi-band compressor, the vocal signal is divided into vocal frequency band signals of a second preset number of frequency bands, the playback audio signal is divided into second playback audio frequency band signals of a second preset number of frequency bands, and it is determined whether the level values ​​of the second preset number of vocal frequency band signals are greater than the second preset value. If so, the second playback audio frequency band signal corresponding to the vocal frequency band signal whose level value is greater than the second preset value is compressed, and a compressed playback signal is output.

[0008] It can be seen from the above scheme that the in-car audio signal is collected by a microphone, and the in-car audio signal is separated into a noise signal, a playback audio feedback signal and a human voice signal, wherein only when the person in the car speaks will there be a human voice signal. When there is a noise signal, the remaining first audio frequency band signals except the noise frequency band are enhanced, and while improving the playback audio signal, the noise signal and the first audio signal do not conflict, ensuring that the increase in volume does not cause auditory discomfort. When someone in the car is talking, the second audio frequency band corresponding to the human voice frequency band is compressed, so that the interference of the music is reduced when the person in the car is talking. Since the in-car audio signal is separated, the sound and noise of the in-car speech can be identified, so that different multi-band compression methods can be adopted, so that while maintaining the in-car music atmosphere, the voice of the person in the car can be heard clearly or the noise does not conflict with the in-car music, thereby improving the overall music experience.

[0009] In a further solution, before determining whether the level values ​​of the first preset number of noise frequency band signals are greater than the first preset value, the following is further performed: detecting the loudness of the noise signal, and obtaining the first preset value according to the loudness of the noise signal.

[0010] It can be seen that the loudness of each noise signal corresponds to a preset value. The first preset value is obtained according to the loudness of the noise signal. The first preset value is obtained through the loudness so that the compressor performs compression according to the loudness. Different preset values ​​ensure that the compression can be dynamically adjusted to enhance the audience's hearing experience.

[0011] In a further scheme, the step of gaining the first playback audio frequency band signal corresponding to the noise frequency band signal less than the first preset value includes: obtaining a first gain compensation value according to the loudness of the noise signal, and gaining the first playback audio frequency band corresponding to the noise frequency band signal less than the first preset value according to the first gain compensation value.

[0012] It can be seen that, since the loudness of a noise signal corresponds to a compression value, the first gain compensation value is obtained according to the loudness of the noise signal, so that the loudness of the music changes with the change of noise, and the volume changes with the change of the environment, thereby enhancing the audience's hearing experience.

[0013] In a further solution, before outputting the vocal signal and the second playback audio signal to the second side chain multi-band compressor, the following is further performed: inputting the noise signal and the first playback audio signal to the first side chain multi-band compressor.

[0014] It can be seen that when someone is speaking, the first played audio signal needs to be amplified according to the noise signal. When someone is speaking, the played audio is not affected by the noise.

[0015] In a further solution, the step of separating the in-car audio signal from the in-car audio signal includes: inputting the in-car audio signal and the currently playing audio signal into a separation algorithm module, and outputting a plurality of separation signals.

[0016] In a further solution, the in-car audio signal and the player's playback audio signal are input into a separation algorithm module, including: separating the playback audio feedback signal from the in-car audio signal according to the current playback audio signal, outputting the remaining signal to be separated, identifying the human voice signal with the remaining signal to be separated, and outputting the noise signal.

[0017] It can be seen that the separation algorithm module includes a separation algorithm, which divides the in-car audio signal into multiple separation signals. Among them, the currently played audio signal is compared with the in-car audio signal, so as to separate the played audio feedback signal from the in-car audio signal, and the human voice signal is identified and separated from the remaining signals to be separated by identifying the human voice signal. If there is no human voice signal, the human voice signal cannot be identified and separated at this time, and the noise signal is directly output.

[0018] In order to achieve the second purpose, the in-vehicle infotainment system provided by the present invention includes: a microphone, a player, a processor and a memory, the microphone is connected to the processor, the player is connected to the processor, the memory stores a computer program, and when the computer program is executed by the processor, the above-mentioned in-vehicle audio adjustment method is implemented.

[0019] In order to achieve the third purpose, the present invention provides a readable storage medium, on which a computer program is stored, and when the computer program is executed, the above-mentioned vehicle audio adjustment method is implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a system structure block diagram of an embodiment of the in-vehicle infotainment system of the present invention.

[0021] Figure 2 It is a flow chart of an embodiment of a method for adjusting vehicle audio of the present invention.

[0022] The present invention is further described below in conjunction with the accompanying drawings and embodiments. DETAILED DESCRIPTION

[0023] The in-car audio adjustment method provided by the present invention separates the sounds in the car, and then uses different multi-band compression methods for different sounds. When there is a noise signal, the remaining audio frequency bands except the noise frequency band are enhanced, so that the noise audio and the first audio do not conflict, ensuring that the increase in volume does not cause auditory discomfort. When someone is talking in the car, the audio frequency band corresponding to the human voice frequency band is compressed, so that the interference of music is reduced when the person in the car is talking, so that while maintaining the music atmosphere in the car, the voice of the person in the car can be heard clearly or the noise and the music in the car do not conflict, thereby improving the overall experience.

[0024] In-vehicle infotainment system example:

[0025] See also Figure 1 The in-vehicle infotainment system of this embodiment includes a processor 1, a microphone 2, a player 3 and a memory 4. The processor 1 can receive the in-vehicle audio signal collected by the microphone 1, and after separating and processing the in-vehicle audio signal, the processor 1 can output the audio signal to the player 3 for playback. The memory 4 stores a computer program, and the processor 1 can execute the computer program to implement the in-vehicle audio adjustment method.

[0026] Embodiment of the method for adjusting the vehicle audio:

[0027] See also Figure 2The in-vehicle audio adjustment method of this embodiment applies the in-vehicle infotainment system of the in-vehicle infotainment system embodiment described above. First, execute step S1 to collect the in-vehicle audio signal, separate the in-vehicle audio signal, and output a plurality of separated signals. Among them, the in-vehicle audio signal is collected by a microphone, and the in-vehicle audio signal is input into the separation algorithm module for separation. The in-vehicle audio signal is a time domain signal, and the in-vehicle audio signal is converted into a frequency domain signal and then input into the separation algorithm module. Among them, the separation algorithm module includes a separation algorithm, and the separation algorithm includes an adaptive filtering algorithm, and the in-vehicle audio signal is compared with the currently played audio signal by the adaptive filtering algorithm, and the playback audio feedback signal is separated, and the remaining signal to be separated is output. The playback audio feedback signal is the audio signal that the player is playing at this time collected by the microphone. The separation algorithm also includes a speech feature extraction technology, and the remaining signal to be separated is identified and separated by the speech feature extraction technology, and a noise signal is output. The speech feature extraction technology can be MFCC speech feature technology or phoneme recognition technology, etc. If there is no voice signal, the human voice signal cannot be identified and separated by the speech feature extraction technology at this time, and the noise signal is directly output. If there is a human voice signal, the human voice signal is identified and separated, and the remaining signal to be separated is a noise signal. By extracting the playback audio feedback signal, the playback audio feedback signal picked up by the microphone is prevented from interfering with subsequent audio analysis. Separating the in-car audio signal into noise signal, playback audio feedback signal and human voice signal allows subsequent sound processing to be performed based on whether there is a human voice or not.

[0028] After outputting a plurality of separated signals, step S2 is executed to determine whether the plurality of separated signals include a human voice signal. If the separated signal does not include a human voice signal, step S4 is executed to input the noise signal and the first playback audio signal into the first side chain multi-band compressor, and a compressed playback signal is output. The first playback audio signal is the audio signal that the player is about to play. Among them, the first side chain multi-band compressor divides the noise signal into noise frequency band signals of a first preset number of frequency bands, such as noise signals of 32 frequency bands or noise signals of 16 frequency bands. If there is no noise signal in the frequency band, the noise signal in the frequency band can be represented by zero. The first playback audio signal is divided into the first playback audio frequency band signals of the first preset number of frequency bands. The loudness of the noise signal is detected. Since a noise signal loudness corresponds to a preset value, and the preset value is a threshold, after the loudness of the noise signal is detected, the first preset value is obtained according to the loudness of the noise signal. It is determined whether the level value of the first preset number of noise frequency band signals is greater than the first preset value. The first preset value is a threshold value. If the level value of a noise frequency band signal is greater than the first preset value, it proves that the signal strength of the noise signal in this frequency band is stronger. Therefore, if the level value of a noise frequency band signal is greater than the first preset value, a first playback audio frequency band signal corresponding to the noise frequency band signal whose level value is less than the first preset value is gain compensated (Make-up Gain) and the gain playback audio signal is output to the player.

[0029] Gain compensation can be performed on the first playback audio frequency band signal corresponding to the frequency band of the noise frequency band signal whose level value is less than the first preset value by adjusting the gain compensation value of the first side chain multi-band compressor so that the first playback audio frequency band signal is gain compensated. Among them, in the step of gaining the first playback audio frequency band signal corresponding to the noise frequency band signal whose level value is less than the first preset value, the first gain compensation value is also obtained according to the loudness of the noise signal. The loudness of a noise signal corresponds to a gain compensation value, and the first gain compensation value is obtained according to the loudness of the noise signal, so that the loudness of the music changes with the change of noise, and the volume changes with the change of the environment, thereby enhancing the audience's hearing. The first side chain multi-band compressor is used to enhance the input signal.

[0030] After the gain playback audio signal is obtained, the gain playback audio signal is also compressed to prevent the gain playback audio signal from having a sharp signal, thereby causing the volume of the gain playback signal to be too loud and damaging hearing, and to ensure the stability of the system.

[0031] In step S2, if the separated signal contains a vocal signal, step S3 is executed to input the vocal signal and the second playback audio signal into the second side chain multi-band compressor, and a compressed playback signal is output. Before the vocal signal and the second playback audio signal are input into the second side chain multi-band compressor, the noise signal and the first playback audio signal are first input into the first side chain multi-band compressor, that is, the second playback audio signal input into the second side chain multi-band compressor is output through the first side chain multi-band compressor. The second side chain multi-band compressor can divide the vocal signal into a vocal frequency band signal of a second preset number of frequency bands, and divide the second playback audio signal into a second playback audio frequency band signal of a second preset number of frequency bands. If there is no vocal frequency band signal in a frequency band, the vocal frequency band signal of the frequency band is represented by zero. After obtaining the vocal signal, the loudness of the vocal signal is detected. Since there is a preset value corresponding to the loudness of a vocal signal, the second preset value is obtained according to the loudness of the vocal signal.

[0032] It is determined whether the level values ​​of the second preset number of human voice frequency band signals are greater than the second preset value, that is, it is determined in which frequency band the human voice frequency band signal has a stronger signal. If the level value of the human voice frequency band signal is greater than the second preset value, the second playback audio frequency band signal corresponding to the human voice frequency band signal greater than the second preset value is compressed, and the compressed playback signal is output to the player. Among them, the second side chain multi-band compressor compresses the human voice signal to reduce the signal strength of the second playback audio frequency band signal corresponding to the frequency band where the human voice frequency band signal is located. Among them, the step of compressing the second playback audio frequency band signal corresponding to the human voice frequency band signal whose level value is greater than the second preset value also includes obtaining a second compression value according to the loudness of the human voice signal, and compressing the second playback audio frequency band signal corresponding to the human voice frequency band signal whose level value is greater than the second preset value according to the second compression value. The second side chain multi-band compressor automatically adjusts the compression ratio according to the loudness of the human voice. If the human voice is detected to be loud, the system will appropriately increase the compression strength to reduce the loudness of the music; when the speech is weak or stops, the system will gradually restore the normal loudness of the music to avoid the influence of volume changes on the listening experience.

[0033] When there is a noise signal, the first audio frequency band signal except the noise frequency band is enhanced, so that the noise signal and the first audio signal do not conflict, ensuring that there is no auditory discomfort when the volume is increased. When someone is talking in the car, the second audio frequency band signal corresponding to the human voice frequency band is compressed, so that the interference of the music when the person in the car is talking is reduced. The sound and noise of the conversation in the car can be separated, so that the voice of the person in the car can be heard clearly while maintaining the music atmosphere in the car, or the noise can be made not to conflict with the music in the car, improving the overall experience.

[0034] Readable storage medium embodiment:

[0035] The method for adjusting the in-vehicle audio in the in-vehicle infotainment system described in the above embodiment can be stored in a readable storage medium in the form of a computer program. When the computer program is executed by a processor, the steps of the embodiment of the method for adjusting the in-vehicle audio in the in-vehicle infotainment system can be completed. The computer readable medium can be a computer readable signal medium or a readable storage medium. The readable storage medium can be, for example, but not limited to: an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0036] The above are only preferred embodiments of the present invention, but the design concept of the invention is not limited thereto. Without departing from the concept of the present invention, more other equivalent embodiments may be included. It is obvious to those skilled in the art that various changes, readjustments and substitutions can be made without departing from the protection scope of the present invention.

Claims

1. A method for adjusting vehicle audio, characterized in that: The method includes: Collecting in-car audio signals, separating the in-car audio signals, and outputting a plurality of separated signals, wherein the plurality of separated signals include a noise signal and a playback audio feedback signal; The step of separating the in-car audio signal comprises: Inputting the in-car audio signal and the currently playing audio signal into a separation algorithm module, and outputting a plurality of the separated signals; Inputting the in-car audio signal and the played audio signal of the player into a separation algorithm module comprises: Separating the playback audio feedback signal from the in-car audio signal according to the currently played audio signal, outputting the remaining signal to be separated, identifying and separating the human voice signal from the remaining signal to be separated by using a speech feature extraction technology, and outputting a noise signal; If there is no human voice signal, directly output the noise signal; If there is the human voice signal, the human voice signal is identified and separated, and the remaining signal to be separated is the noise signal; Among them, only when the person in the car speaks will there be the human voice signal; Determine whether the plurality of separated signals contain the vocal signal; if the separated signals do not contain the vocal signal, input the noise signal and the first playback audio signal into a first side chain multi-band compressor, wherein the noise signal is divided into noise frequency band signals of a first preset number of frequency bands, and the first playback audio signal is divided into first playback audio frequency band signals of a first preset number of frequency bands, and determine whether the level values ​​of the first preset number of noise frequency band signals are greater than a first preset value; if so, perform gain compensation on the first playback audio frequency band signal corresponding to the noise frequency band signal less than the first preset value, and output the gain playback audio signal; If the separated signal contains the vocal signal, the vocal signal and the second playback audio signal are input into a second side chain multi-band compressor, wherein the vocal signal is divided into vocal frequency band signals of a second preset number of frequency bands, and the second playback audio signal is divided into second playback audio frequency band signals of a second preset number of frequency bands, and it is determined whether the level values ​​of the second preset number of vocal frequency band signals are greater than a second preset value. If so, the second playback audio frequency band signal corresponding to the vocal frequency band signal greater than the second preset value is compressed, and a compressed playback signal is output.

2. The method for adjusting the in-vehicle audio according to claim 1, characterized in that: Before determining whether the level values ​​of the first preset number of noise frequency band signals are greater than a first preset value, further executing: The loudness of the noise signal is detected, and a first preset value is obtained according to the loudness of the noise signal.

3. The method for adjusting the in-vehicle audio according to claim 2, characterized in that: The step of gaining the first playback audio frequency band signal corresponding to the noise frequency band signal less than the first preset value includes: A first gain compensation value is obtained according to the loudness of the noise signal, and a first playback audio frequency band signal corresponding to a noise frequency band signal that is smaller than the first preset value is gain-enhanced according to the first gain compensation value.

4. The method for adjusting the in-vehicle audio according to any one of claims 1 to 3, characterized in that: Before outputting the vocal signal and the second playback audio signal to the second side chain multi-band compressor, the following is further performed: The noise signal and the first playback audio signal are input into a first side-chain multi-band compressor.

5. In-vehicle infotainment system, including: A microphone, a player, a processor and a memory, wherein the microphone is connected to the processor, the player is connected to the processor, and the memory stores a computer program, wherein the computer program, when executed by the processor, implements the vehicle audio adjustment method as described in any one of claims 1 to 4.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed, the vehicle audio adjustment method according to any one of claims 1 to 4 is implemented.

Images