Audio Signal Processing Solutions
ADI LISTN—Hear What Matters
Analog Devices understands that an immersive, in-cabin, audio experience is one of the primary differentiators in purchasing a car. From infotainment to hands-free communication, with a person or with a machine (think “Siri”, “Alexa,” or “OK Google”), ADI has utilized its sophisticated IP to develop ADI LISTN™—a portfolio of high performance, microphone audio signal processing software solutions that fulfill the highest automotive industry standards.
With a matured IP equipped on millions of cars on the road, this dependable software solution is easy to integrate and tune, and thus gets you to market faster.
With ADI’s acquisition of Ruwisch & Kollegen GmbH, we’ve enhanced our portfolio of audio signal processing software solutions immensely. All ADI LISTN microphone audio signal processing software solutions are available as building blocks in SigmaStudio® and as libraries for integration with CrossCore® Embedded Studio™ (CCES), on ADI SHARC® DSPs, or on any tool chain as required for compiling on a third-party CPU or SoC.
ADI LISTN High Performance Benefits Include:
- Industry-leading software that is flexible, customizable, and supports any hardware platform.
- Enabling hardware implementations that are resource efficient and optimized, resulting in high quality audio signal processing and deterministic performance with low latency.
- Enabling compliance with the highest quality automotive standards (ITU-T p.1100/1110/1120/1140) that pass multiple Apple CarPlay and Google Android Auto certifications.
- Simple-to-use tuning tool that cuts the evaluation time compared to other existing solutions, for a fast time-to-market schedule.
- Over 20 patents filed—ADI LISTN is uniquely positioned to provide exceptional performance not only in human communication but also when processing audio before the recognition engine receives it to minimize error rate and deliver a great in-cabin experience.
Voice Recognition
All aspects of voice recognition require high quality audio processing for satisfactory performance in all use cases (high velocity, music playback, passenger conversations, etc.). ADI’s state-of-the-art solutions, in conjunction with the audio preprocessing (AEC/NR/BF) algorithm, deliver the perfect product for all range of automotive customers.
ADI’s wake-up word solution for voice recognition is part of ADI LISTN and supports the latest industry standards for automotive infotainment systems.
In-Car Communication (ICC)
One direct benefit of ADI’s sophisticated IP is In-Car Communication. ICC is a voice-band, audio use case where the processed microphone signal is not sent to a distant receiving entity but is output on loudspeakers within the vehicle. This requires very low latency signal processing, otherwise the loudspeaker output would be perceived as an echo of the speaker’s own voice.
The closed-loop situation bears the risk of feedback problems known as howling or ringing, which must be kept under control by the ICC algorithm. ICC is most important in front-to-rear (unidirectional) communication, so the driver can easily speak to rear passengers without turning their head. But it can also be implemented in a bidirectional manner adding rear-to-front support.
A derivative of ICC is in-car karaoke, which can be understood as front-to-front ICC (because the driver wants to hear themself singing!). In-car karaoke is usually preferred louder than ICC, as ideal ICC is barely noticed by passengers but missed when turned off.
Audio Signal Processing Software Algorithms
Acoustic Echo Cancelling (AEC), Noise Reduction, and Beamforming are the main algorithmic components of the ADI LISTN audio signal processing software suite. These technologies are the foundation of ADI’s voice processing solutions, which manufacturers require in a truly immersive, in-cabin experience.
Acoustic Echo Cancelling (AEC)/Noise Reduction Technologies
Different disturbances interfere with the speaker’s voice signal and without elaborate signal processing, communication would not be possible, neither between humans nor with a machine.
AEC and NR control the two disturbing signal components in a vehicle: loudspeaker-output and driving-noise. Since the source signals of loudspeaker outputs are known, these disturbances can be completely removed from the microphone signal via a two-step process.
Linear echo cancelling is an adaptive algorithm that calculates the impulse responses between loudspeakers and microphones, and estimated echoes are calculated that are then subtracted from the microphone signals. Nonlinear residual echo suppression removes the echoes that remain after linear cancelling, as the human ear is extremely sensitive to echoes of one’s own voice.
For other omnipresent disturbances (driving noise), it can be assumed that this changes slowly over time in a running car and an adaptively calculated noise estimation serves as internal representation of the actual, current noise that is then reduced by the NR algorithm by up to 20 dB.
Beamforming Technology
Beamforming starts with two or more microphones arranged as an array. Spatial information of the sound in the vehicle is considered and it is thus possible to distinguish different directions of sound impact.
With beamforming, you can focus on a certain seat position and reduce other voices in the car. This helps with voice recognition because other interfering voices can substantially degrade voice recognition performance. For human communication, beamforming is useful as an add-on to noise reduction because it can reduce non-stationary noise components if they come from off-beam directions.
Listen to a demonstration of our voice processing technology.
ADI LISTN runs not only on ADI SHARC and SigmaDSP processors, but also on third parties’ SoCs—giving customers the freedom to choose their preferred system architecture.
When used in combination with ADI’s A2B® technology, ADI LISTN leverages time synchronicity by eliminating signal clock regeneration through the daisy chain—ultimately reducing overall BOM cost and simplifying microphone wiring schemes to enable a full turn-key solution.
Audio Signal Processing Software Tools
ADI LISTN Tuning Tool
The ADI LISTN tuning tool is a graphical user interface (GUI) that connects to the target via Ethernet, UART, or SPI. With the ADI LISTN automotive audio signal processing suite and its tuning capabilities, it is easy to create a system that passes the relevant automotive specifications such as ITU-T p.1100/1110/1120/1140 and Apple CarPlay.
SigmaStudio
The SigmaStudio® graphical development tool is the programming, development, and tuning software for the SigmaDSP® and SHARC audio processors and A2B transceivers. Familiar audio processing blocks can be wired together as in a schematic, and the compiler generates DSP-ready code and a control surface for setting and tuning parameters.
CrossCore® Embedded Studio (CCES)
CrossCore Embedded Studio is a world-class integrated development environment (IDE) for the ADI Blackfin®, SHARC, and Arm® processor families. CCES also offers Blackfin and SHARC developers highly integrated add-in support for drivers, services, and algorithmic software modules.