Acoustic experience is becoming a new dimension that defines the value of automobiles. Audio has rapidly become not only a means of communication and entertainment, but also a way to improve people's lives.
Over the past decade, car consumers' expectations for audio systems have risen significantly. A well-performing audio system, clearer voice interaction, and customized and personalized content help enhance the user experience of car owners, which has also been widely recognized by many Oems and the market.
    The positioning of cabin audio in new energy vehicle products is undergoing a transformation, evolving from a single driving function in the past to an entertainment space, family space, and private space that meets individual needs.
    The key sign of this transformation is the comprehensive upgrade of user experience: in terms of entertainment sound effects, we are no longer content with simple stereo effects such as radios and Bluetooth, but pursue advanced surround sound, spatial audio and independent sound zones and other technologies to significantly enhance the subjective feelings of drivers; In terms of interaction, the cabin environment can be controlled through voice, routes can be planned, and local communication compensation within the cabin can be achieved. In terms of NVH experience, actively controlling engine and tire noise enables automakers to more easily optimize the materials and controls of the body and chassis. Meanwhile, the ever-growing demand for in-vehicle audio and autonomous driving in new energy vehicles essentially places higher demands on the performance and computing power of on-board chips. The improvement in performance and computing power has enabled functions that were previously impossible to complete through real-time computing to now be realized at a low cost and mass-produced.
   The key to improving audio quality lies in DSP. Nowadays, high-end DSP is still dominated by international giants, but domestic manufacturers are also catching up vigorously and striving to break the foreign monopoly.

Modern cockpit system composition
    The cockpit electronic solutions mainly include the main unit, audio power amplifier, audio bus, etc. Among them, the host system integrates multiple functions such as audio/video playback, navigation, driver assistance and in-vehicle communication, aiming to provide drivers and passengers with a comfortable and convenient driving and riding experience. The typical functions of the system include FM/AM/ digital/satellite broadcasting, CD/DVD playback, multimedia peripheral access, rear-seat entertainment, navigation, camera integration, gesture recognition, voice recognition, Bluetooth connection and wireless communication connection, etc.
    The independent audio power amplifier is connected to multi-channel speakers to create various audio environments inside the vehicle and provide high-fidelity audio enjoyment. The typical functions of the system include powerful audio processing, active noise cancellation in the cabin, and engine sound effect simulation.
    The audio bus is used to transmit audio and control signals between modules such as the host, power amplifier, active noise cancellation module, and microphone/microphone array.
    In addition to the basic audio/video playback functions, modern in-vehicle entertainment systems have put forward higher requirements for audio/video performance, connection performance, human-machine interface, operational convenience and power consumption. The system has introduced more and more efficient digital signal processing technologies and integrated more advanced audio algorithms to enhance the acoustic experience or compensate for the acoustic response in the cockpit.

ADI: Active noise cancellation becomes a new demand
    Since the tram has no engine, its noise level has been greatly reduced, which makes other noises more prominent. Therefore, active noise reduction technology is becoming increasingly important.
Previously, most car manufacturers used passive sound-absorbing materials to help reduce noise. However, this will increase the vehicle's weight and thereby raise fuel consumption. So, manufacturers drew on the active noise cancellation (ANC) technology used in high-end audio noise-cancelling headphones and deployed this technology in millions of vehicles. Although ANC can effectively eliminate the narrowband engine noise of vehicles, it has little effect on suppressing the broadband noise (road noise) generated by tires.
Teams from Analog Devices (ADI) and Hyundai Motor Corporation (HKMC) have joined forces to deploy the industry's first all-digital road noise active noise control (RANC) system. The challenge they face is how to implement RANC in a practical and easy-to-produce way. At the beginning of 2020, the pioneering RANC system was accelerated into mass production and applied to Hyundai's flagship product, the GV80 car. Taking advantage of the new silent cabin, Hyundai Motor has deployed the Lexicon® premium audio system with Quantum Logic®Surround technology to create a stage-level audio effect, which has been optimized and tuned for road conditions. The GV80 is equipped with 21 speakers and subwoofers, which are ingeniously distributed at 17 positions in the cabin. Combined with a 1050-watt amplification power, it creates a perfect immersive listening experience for each passenger. Although the commuting time to work in the future may still be long, this collaboration between ADI and Hyundai Motor will help drivers and passengers relax in a safer, quieter and more entertaining personalized space.
    Hyundai Motor is looking to forge a path for the future - going beyond noise reduction technology to make its entire automotive product line take into account the feelings of both drivers and passengers, providing a more immersive, entertaining and high-quality audio experience.
    In the past, to reduce noise inside the vehicle and make the car feel more upscale, car manufacturers would add sound barriers at all possible locations (such as the floor, fairing, etc.). After the advent of lightweight materials, they only added a lot of inhibitory effects and then affected fuel efficiency due to the additional weight. Therefore, they turned to a simple simulation technology - active noise cancellation, which was initially used in the 1950s to reduce the deafening engine sounds in helicopter and aircraft cabins and was later widely adopted in the headphone industry.
    ANC is a simple and excellent solution that can eliminate narrowband noise (not just suppress it). Just as ripples in a lake combine with troughs, the function of ANC is to flatten sound waves or cancel them out.
Eliminating road noise is not actually aimed at achieving quietness, but rather at preventing noise from causing fatigue to drivers and passengers.

The working principle of active noise cancellation
    The working principle of ANC technology is the same as that of noise-cancelling headphones, using software and components such as DSP and microphones to measure and reduce noise. The in-car microphone picks up the engine noise and transmits it to the digital signal processor (DSP). The ANC algorithm processes it in the DSP and then sends back the reverse audio signal through the speaker. The driver and passengers find it hard to detect the sound produced by this destructive disturbance, but it is sufficient to counteract the engine noise.
The ANC app "listens" for unwanted external engine noise and then plays a sound that is 180 degrees out of phase. In this way, the noise is "eliminated" before it reaches your ears. Digital processing must be extremely fast to complete this task.
    ANC technology is only effective against predictable constant low-frequency noise, such as narrowband noise from engine rumbling. Randomly occurring broadband noise (such as pits) will seep in. Because the signal and computing speed of ANC are not sufficient to respond quickly and generate reverse waves. It is difficult for its software and hardware to complete the noise reduction process within 4 ms (4 milliseconds). This is the time it takes for road noise to reach passengers' ears. Modern automobiles are seeking better solutions.
Of course, this disadvantage can sometimes become an advantage. It is very important to be able to hear the horns of cars or the sirens of fire engines/ambulances when driving. Since ANC only offsets continuous low-frequency noise, these important unconventional high-frequency cues will come loudly and clearly.
As early as 2012, Analog Devices (ADI) provided Hyundai Motor with digital signal processing (DSP) technology suitable for its advanced audio amplifiers. In the following years, this partnership rapidly developed into a focus on road noise active noise control (RANC) systems for modern automobiles.
    RANC technology not only has low signal latency and fast processing speed, but also travels more than four times faster than noise from the road to the cabin. In just 2 ms (2 milliseconds), RANC analyzes road noise and sends reverse sound waves through the car's stereo speakers to cancel it out before anyone inside the vehicle hears it. In tests, RANC can reduce cabin noise by half or 3 dB (3 decibels).
    The RANC system proposed by Hyundai Motor consists of four main components: an accelerometer, a processor, a microphone, and a speaker.
Accelerometer: The accelerometer is located near each tire and can detect the vibration of the tires and suspension.
The processor: The DSP proprietary algorithm of the adaptive controller offsets the vibration noise within a few milliseconds through the equal-amplitude reverse sound waves transmitted by the speaker.
Microphone: Meanwhile, the microphone measures the sound inside the cabin to ensure the generation of an accurate noise-cancelling signal.
The speakers in the cabin will turn off the signal circuit, and the interfering sound waves will also be reduced. Therefore, the internal area around each passenger is quieter, and the auditory experience during driving is more comfortable.
    From 2015 to 2016, Hyundai Motor and ADI have been collaborating on this solution. In early 2020, two teams successfully deployed the industry's first all-digital road noise reduction technology product on Hyundai Motor's GV80.
    Vlad Bulavsky, general Manager of the Automotive Cockpit Connectivity Division of ADI, said: "RANC is a very good application example, and various different technologies need to be used to achieve efficient system solutions." ADI's DSP, A2B connectivity, accelerometer and software IP are seamlessly integrated to provide high-performance and cost-effective solutions for highly challenging acoustic problems.
    In addition, Bose has also joined hands with ADI to leverage advanced automotive audio technology to help control road noise and provide a more immersive and relaxed cabin experience.
    Bose employs ADI's breakthrough technology in automotive audio system design to create an immersive and high-quality auditory experience inside the cabin, including the outstanding QuietComfort® road noise control. Bose found that traditional mechanical noise reduction methods reduce low-frequency noise from the road, tires and chassis, which requires the use of thick sound insulation materials, primers and heavier chassis components, and this will increase the weight of the vehicle body. The increased mass will have a negative impact on aspects such as fuel economy and the driving range of electric vehicles.
    Bose's road noise control system employs ADI's DSP technology and uses an accelerometer and microphone connected via A2B to detect unnecessary cabin noise. This system can quickly process noise data, then generate noise reduction signals and emit them through the vehicle's speakers. These signals are precisely matched with road noise, ultimately effectively reducing noise and meeting the vehicle manufacturer's requirements. Bose's unique algorithms, combined with ADI's advanced processing capabilities, ensure a quieter and more immersive cabin experience. This is particularly beneficial for electric vehicles, as they cannot use the sound of traditional engines to mask the noise from tires and roads.
    In the process of creating an outstanding automotive cabin audio experience, ADI found that emerging vehicle computing solutions are difficult to support the real-time (low-latency) requirements of complex audio applications such as road noise reduction. For this reason, Bose began to use ADI's SHARC audio processors and A2B to provide high-performance audio solutions for its Bose Active Sound Management (ASM) product line.
After adopting advanced noise reduction and beamforming technologies, each passenger in the vehicle can enjoy their own personalized audio micro-environment.
    Bose SeatCentric technology enables each passenger in the car cabin to choose their own listening experience without interfering with other passengers. For instance, the father can answer work calls or listen to navigation instructions at any time, while other family members can enjoy music at any volume at the same time, and all these operations can be adjusted from each seat. In addition, the rear seat audio can be actively cancelled to facilitate the rear seat passengers to take a short rest or listen to the audio on their own devices, while the front seat passengers will not be affected. All of this is attributed to SeatCentric technology.
    The A2B and SHARC audio processors employ the same ADI underlying technology, and the low-latency audio provided by these technologies is crucial for realizing Bose's vision of a personalized SeatCentric experience.
Overall, after adopting advanced noise reduction and beamforming technologies, each passenger in the vehicle will have their own personalized audio environment, and the vehicle will also become a personal digital assistant.

Overview of ADI's audio solutions:
    ADI offers a wide range of video decoding converters/encoding converters/coprocessors with outstanding signal performance and functionality. It can also provide complete low-power LVDS/HDMI®/MHL/APIX2 interface devices.
    The audio bus provided by ADI - automotive audio bus (A2B®), is mainly used for the connection between various audio modules. It can simultaneously transmit audio and control signals with a pair of twisted wires and also provide power.
    ADI's fixed-point audio DSPS, SigmaDSP®, floating-point audio DSPS, SHARC®, and multi-core floating-point audio SOC Griffin are widely used as high-performance/high-efficiency audio post-processing and dedicated audio processors for routing/mixing in hosts and independent audio amplifiers. The award-winning graphical user interface tool SigmaStudio™ makes DSP programming as simple as drawing a flowchart.
    ADI's highly integrated Blackfin® DSP series is suitable for multimedia connectivity, network connectivity and software audio processing applications.
In addition to DSP, a wide range of software modules are also provided, such as decoders, surround sound, virtualization, echo cancellation, call noise reduction and active noise reduction, etc.
The audio codec converters provided by ADI offer a wide range of channel options and feature high SNR/THD.
    ADI's multi-functional signal conditioning and photometric sensors integrate AFE, ADC, LED drivers and timing cores, and can be used in a variety of optical measurement applications, such as close-range detection in the cockpit and gesture recognition.

Texas Instruments: The New Generation of DSP redefines Infotainment systems
   To meet the demands of original equipment manufacturers and end consumers, Texas Instruments (TI) has launched a new generation of audio processors. They have a higher degree of integration and stronger computing power, which helps simplify the design of audio systems and improve the cost-effectiveness of the systems.
   The new generation of products includes the AM62D-Q1 processor and the AM275x-Q1 MCU, integrating TI's C7x DSP core, memory and other components. This not only ensures the high performance of the system, but also provides a margin to support future upgrade requirements.

The specific highlights are as follows： 
· A powerful heterogeneous processor that supports AI algorithms
    The AM62D-Q1 is equipped with a 4-core Cortex-A53 processor, a single-core Arm Cortex-R5F, and a C7x DSP with a matrix multiplication accelerator. The AM275x-Q1 integrates dual-core or quad-core Arm Cortex-R5F and C7x DSP.
    It is particularly worth noting that the new generation C7x DSP core supports a 256-bit vector operation architecture. The core can access L2 memory in a single cycle, which is faster. At the same time, the main frequency has reached 1GHz, providing 40GFLOPS of computing power, and the performance of its matrix multiplication accelerator (MMA) is as high as 2TOPS. The C7x core combined with the matrix multiplication accelerator forms an on-chip neural processing unit.
    With the help of AI accelerators, not only can general audio algorithms be run, but also various types of AI algorithms can be supported. Compared with the traditional DSP architecture based on scalars, the DSP core based on vector architecture has higher audio processing performance and excellent scalability, performing well in both low-end and high-end audio processing.

· Larger capacity storage
    The demand for computing power in advanced real-time audio is not only reflected in the CPU but also in the memory aspect. To achieve high-throughput audio processing, the functional units of the DSP core need to access memory in each cycle. The AM2754 is a DDR-free MCU that features up to 4.5MB of single-cycle access L2 memory and up to 6MB of L3 memory. The AM62D-Q1 is a DDR-based processor, suitable for high-quality audio designs that require high-speed external memory. The AM62D-Q1 features 1.25MB of single-cycle access L2 memory and a 32-bit LPDDR4 controller for additional high-speed external memory.
    · Rich interfaces and peripherals
     The key to an audio transmission network is precise and low-latency communication and synchronization. The Ethernet Audio and Video Bridging (AVB) standard is an ideal choice because there is already an Ethernet in the vehicle for connecting other ECUs, which can simplify the wiring architecture and reduce the weight and cost of system cables.
     TI's new audio DSP integrates an Ethernet switch and supports TSN, making it highly suitable for modern audio data transmission applications requiring high speed and high bandwidth.
     In addition, in terms of audio interfaces, TI also supports multiple multi-channel audio serial ports (McASP), which facilitates system expansion.

·Hardware security module

     Nowadays, the safety of automobiles is becoming increasingly important. To prevent hacker intrusions, various safety protection modules are needed. TI's processors support secure boot, TEE, and various encryption algorithms, especially the national cryptography SM3 and SM4, meeting the information security requirements of vehicles. In terms of functional safety, the system capability meets the ASIL-D target, the hardware integrity meets the ASIL-B target, and complies with AEC-Q100.

    The HSM module built into TI DSP integrates secure storage, encrypted hardware acceleration, secure CPU and hardware interfaces connected to the rest of the system, conforming to the standards of Secure Hardware Extension 1.1 and EVITA (Electronic Secure Vehicle Intrusion Protection Application).
    Platformization strategies for two architectures
    TI offers both the DDR-based AM62D and the AM275x without DDR, allowing customers to flexibly choose based on their own cost and computing power requirements. By selecting different products, as the same DSP and Arm architecture are adopted, customers' algorithm investment will not be wasted.

One-stop audio solution

    For automotive audio systems, TI offers a complete range of embedded, analog and power products and solutions, providing customers with one-stop solutions.
    TI's amplifiers, data converters and MPUs can enhance the low noise, distortion and latency required for the in-car experience, thus ensuring high-quality sound. The amplifier features advanced protection characteristics such as overcurrent protection, real-time IV detection, load diagnosis, and powerful ESD and EMI protection.
    Meanwhile, in terms of reducing battery consumption and enhancing the driving range, the advanced modulation scheme of TI's Class D amplifiers, when combined with the power tracking function of the boost converter, can achieve better power efficiency.
    In addition, TI also offers customized reference designs, training and support for terminal devices, thereby reducing the development difficulty and accelerating the system development cycle.

AKM: ANC is achieved by using accelerometers and DSP

    Road noise elimination technology is used to reduce the noise caused by the vibration of the road surface during vehicle operation. This technology achieves noise reduction by performing low-latency real-time calculations on the input signals from the acceleration sensor and constructing a secondary sound field inside the vehicle.
    AKM offers ultra-low latency ADC chips that can quickly transmit sensor signals to AKM DSP chips equipped with Silentium road noise cancellation algorithms. By combining software and hardware, it creates a unique low-latency in-vehicle ANC (Active Noise Cancellation) solution.
    The most direct way to deal with changes in the audio environment is for users to manually adjust the volume. However, frequently adjusting the volume in a constantly changing road environment can significantly affect the user experience. In response to this issue, typical existing technologies - speed-linked volume adjustment and ANC - although they contribute to the control of in-car audio systems, each still poses challenges.

Domestic manufacturers are catching up vigorously
    While international manufacturers are upgrading DSPS, many domestic chip manufacturers are also competing in the audio market.
Rockchip
    The Rockchip RK2118 chip features outstanding hardware capabilities and performance. It adopts high-bandwidth HiFi 4 DSP and large-capacity SRAM, and integrates Rockchip's self-developed audio NPU, achieving a deep integration of high-performance "multi-core heterogeneous" technology architectures. It can support AI algorithms such as vocal separation/enhancement, music separation, noise reduction, and virtual surround sound.
    RK2118 adopts high-bandwidth HiFi 4 DSP and large-capacity SRAM, and integrates Rockchip's self-developed audio NPU, achieving a deep integration of high-performance "multi-core heterogeneous" technology architecture. It supports automotive algorithms such as vocal separation/enhancement, music separation, noise reduction, ECNR, and virtual surround sound. RK2118 has aroused great interest in the automotive industry. It is worth noting that it has been adopted by leading automakers and Tier 1, which is sufficient to demonstrate that the RK2118 can meet the high standards of various passenger car audio applications.
    With the advancement of mass production, RK2118 will have a significant impact on the automotive and consumer electronics markets. Its advanced audio processing capabilities, combined with the reliability and efficiency of HiFi 4 DSP, will be able to adapt to the audio demands of multiple scenarios in vehicles, significantly enhancing the driving and riding entertainment experience." Henry Huang, the audio marketing director of Rockchip Microelectronics, said.
    The product was put into mass production in the fourth quarter of 2024. This cutting-edge SoC is expected to leverage the powerful capabilities of HiFi 4 DSP and its extensive ecosystem to effectively enhance the performance and experience of automotive audio systems and traditional consumer electronic audio products.
    The reason for choosing HiFi 4 is that its performance has been greatly improved compared with the previous generation HiFi 3. The HiFi 4 DSP features advanced configuration options, including a vector floating-point unit (VFPU), which significantly enhances its ability to handle complex audio processing tasks. Therefore, RK2118 becomes an ideal choice for automotive audio processing applications that strictly require high-fidelity sound quality and low latency, as well as for traditional consumer electronic audio products.

Goodix Technology
    Starting from the deep integration of software and hardware, Goodix Technology has been dedicated to laying out and creating an overall automotive audio solution that integrates hardware such as in-vehicle audio amplifiers and software such as voice processing, audio effect processing, and active noise cancellation, to facilitate a richer, smarter, and more enjoyable advanced driving experience. Among them, the audio software solution, with its flexible and reliable integration method, supports mainstream SoCs and various DSP platforms, effectively reducing system costs. One-stop solutions better help customers shorten the project development cycle. At present, the CarVoice solution has been successfully commercialized by many mainstream models of brands such as Hyundai, Nissan, Honda, Toyota, Chery, and GAC, as well as in-vehicle infotainment solution providers, thanks to its advantages of low resource consumption and stable and reliable performance.
Guoxin Technology
   To meet the urgent demand of a wide range of passenger car suppliers and Oems for upgrading cabin audio systems, Guoxin Technology has recently successfully developed a DSP chip - CCD5001 for high-end cabin audio processing in the new energy vehicle market, and has planned a complete series of products. The CCD5001 chip product is a high-performance DSP chip developed based on the HIFI 5 architecture core. It is specifically designed for active noise control, high-end surround sound effects, intelligent voice interaction and other application scenarios in automotive platforms. While meeting the requirements of extremely low latency, high floating-point performance and multi-channel signal processing, the chip can also be widely applied in industrial, transportation and other fields. Meet the requirements of high-reliability signal processing or real-time control.

    In audio DSP tasks such as FIR/IIR/FFT, the CCD5001 uses dedicated hardware accelerators to execute general algorithms, such as finite and infinite impulse response (FIR and IIR) filtering, allowing real-time audio filtering outside the core DSP, thereby enhancing the effective performance of the core and providing up to 2.5 times the performance compared to similar chips. The chip is equipped with a rich array of audio interfaces (SPORT, SPDIF) and built-in audio subsystems (DAI, PCG), capable of achieving parallel input and output functions for 8 TDM32 channels, totaling 256 channels. It supports series topology, significantly simplifying the connection of in-vehicle wiring harnesses and reducing system costs. The flexible data path configuration and high data transmission efficiency of the chip cover the expansion requirements of audio systems from low-end to high-end. The rich peripheral control resources enable the CCD5001 to be conveniently integrated into the hardware platforms of various audio management systems. In terms of driver software, CCD5001 fully respects the habits of developers and users. Suppliers and Oems can extensively reuse the drivers and firmware of existing architectures, reducing development costs. In terms of application software, CCD5001 plans to provide customers with industry-standard tuning tools, achieving a perfect balance between development quality and efficiency.
    With the accelerated penetration of new energy vehicles in the passenger car market, the sudden rise of new energy vehicles has driven a comprehensive upgrade of in-vehicle cabin audio systems, and the value per vehicle has increased rapidly.
    In terms of in-cabin audio computing power, foreign giants have long monopolized it, and the domestic production rate is low. The supply chain crisis was particularly prominent during the epidemic. As the first automotive-grade audio processing DSP chip of Guoxin Technology, CCD5001 has laid a solid foundation for the future layout of DSP series chips.
   The company not only pursues breakthroughs in high-end cabin audio applications, but also plans to launch a series of mid-to-low-end DSP products in the future. These series of products will deeply study the audio processing requirements of different vehicle models, and ultimately help customers enhance the competitiveness of domestic automobiles by optimizing costs and ease of use.