Except in cases where they are being flambéed by unscrupulous financial types, numbers can give you a pretty accurate assessment of the health of a business or market sector. Consider Intelligent Vehicle Infotainment Systems (IVI), for example.  Earlier this year the technology market intelligence company ABI Research provided the best evidence yet that multimedia information consoles are now the pre-dominant feature of dashboards in new cars; the firm predicted that shipments of connected automotive infotainment systems would grow from 9 million in 2013 to more than 62 million in 2018—a nearly 7 fold increase-- with connected navigation, multimedia streaming, social media, high resolution displays and in-car Wi-Fi hotspots among the key reasons why.

But while there’s no questioning the attractiveness of seamlessly keeping up to date with your life while motoring down the highway, if there is a potential fly in the ointment it is this: with 90% of accidents said to be caused by driver error and, even without knobs (today’s cars come with touch screens and flat panel buttons), to utilize the display drivers will take their eyes off of the road for up to two seconds, so driver distraction concerns are becoming an issue.

One potential answer is speech recognition. Until recently car-based, voice-response systems were used mainly for tasks such as automated dialing of hands-free Bluetooth devices or to operate the basic functions of radio/music head units. Going forward, however, voice based control is poised to make the leap from purely task-oriented command and control to a more sophisticated, user-centric interface designed to negate distraction issues at a time when the social media interactivity phenomenon is quickly making its way into the automotive cockpit.

Speech-recognition in the form of hands-free text-to-voice or voice-to-text operations are delivering greater accuracy and employing more flexible grammar libraries, which allow users to interact in a more conversational way instead of being restricted to fixed, predefined menu phrases. For example, Ford's SYNC, based on the Microsoft's Windows CE operating system, supports up to 10,000 voice commands with no training required for the system to recognize voice commands.

Speech based IVI applications require more computing performance to increase accuracy and improve UIs. Developers can meet these challenges using the Intel® Atom™ processor and Intel Software Development products. Improved microphone performance is one method to enhance voice recognition systems. Today’s microphones offer 61dB SNR and the industry is quickly adopting 63DB as its standard (employing three MEMS microphones) with 70DB not far off. MEMS microphones are behind much of this advance; they produce an electrical signal from a change in capacitance caused by the movement of a membrane relative to a stationary plate. A MEMS acoustic sensor uses a vibrating membrane and a fixed plate to convert acoustic pressure waves into this change in capacitance. A CMOS interface IC then converts the change in capacitance into a digital PDM modulated output (Fig. 1).


Figure 1 Basic operation of a MEMS mic (Source: IDF 2013).

From an engineering perspective the different microphone types and placements pose design challenges to enabling natural speech because microphone configuration can vary dramatically both in type and number. It has been demonstrated that 3 and 4 mics used in similar environments can improve speech recognition performance by 10% or more compared to one or 2 mic set-ups.

When using more microphones, the signal processing workload increases for the IVI system with the average normalized latency increasing about linearly with the number of microphone channels. Happily, the Intel® Atom™ processor family E3800 (formerly code named Bay Trail)  is capable of handling inputs from multiple sources simultaneously, with low CPU overhead and without using a DSP, which can lower system cost. The processor also is industrial grade (-40o to +85oC), satisfying automotive requirements. No fan is needed since the processors in the family deliver their performance in a sub-10-watt system on a chip.

If more DSP functionality is needed for audio apps the Intel Haswell microarchitecture is the first Intel product with an Integrated DSP core dedicated to audio (an180MHZ core, 200KB SRAM and  I2C I2S audio codecs are used). I n this way developers can deliver a differentiated user experience and still have computing headroom on-board for future acoustic improvements or to run other applications concurrently.

On the visual side the Intel Atom processor E3800 product family delivers enhanced media and graphics capabilities at low power. These SoCs are based on the Silvermont microarchitecture, utilizing Intel’s industry-leading 22nm process technology with 3-D Tri-Gate transistors. This new microarchitecture delivers significant improvements in computational performance and energy efficiency over previous-generation CPU microarchitectures. It includes a new out-of-order execution engine for superior compute performance, outstanding power management capabilities, and enhanced security. Intel® Virtualization Technology increases virtualization performance by allowing the operating system more direct access to the hardware.

But getting back to Intel Atom processor, using the new Bay Trail product can enhance animated text and graphics, facilitate camera use (increasingly popular in vehicles) and improve visual textures,  transitions and other effects that are all glued together to achieve a truly immersive in-car experience. Other benefits include faster media conversions, stereoscopic 3D capabilities, highly efficient image processing, and enhanced HD video transcoding over HDMI with Intel® Gen 7 graphics.

Other innovative and original automotive solutions to control in-car infotainment consoles are being made possible through the Tizen OS development platform for IVI embedded computing systems. Intel is a key contributor and part of the Technical Steering Group of Tizen. Last week at CEATEC, the IT and Electronics exhibition and conference held at the Makuhari Messe complex just north of Tokyo), Intel booth drew attention via a demo in which Systenahttp://www.systena.co.jp/eng/(Tokyo, Japan) showed off a unique software-based Human Machine Interface (HMI) for Tizen IVI that can change the look of the HMI display to suit the user’s taste and personality (Fig.2). The solution utilizes information provided by the driver’s smartphone or other handheld device which is recognized through Near Field Communication (NFC).



Figure 2 Systena’s Tizen-based HMI can vary according to user’s preference.


Also at the Intel booth OBIGO of Korea displayed an HTML5 browser for automotive use employing Tizen IVI that not only showed infotainment features on screen-- such as the ability to access e-mail-- but also provided vehicle data such as air conditioning settings.

An IVI Telematics computer using Intel® Atom™ processor E3827


Figure 3 The Nexcom VTC 1010

Nexcom describes its VTC Series as “the Intelligent Vehicle Telematics Computer for a Mobile World.” A quick look at its features and specs shows why. The Nexcom VTC 1010 fanless In-Vehicle Computer (Fig. 3 above) features the next generation Intel Atom processor E3827 at 1.75GHz, with powerful graphic and multimedia enhancement. The unit comes in a 1 DIN compact enclosure with built-in CAN BUS 2.0B interface and optional OBDII (ASE J1939/ J1708) port to monitor the vehicle’s operating status real-time. VTC 1010 can be configured to work with two independent WWAN connections and can effectively increase bandwidth for faster data transfer over the air. VTC 1010 also supports two-way voice communication. Equipped with intelligent vehicle power management, VTC 1010 can be woken up by ignition, timer, or remote dial-up for flexible operation.


For in-vehicle infotainment applications the VTC 1010 boasts a built-In G-sensor, Gyroscope, and e-Compass sensors. A built-In Ublox 6 GPS comes with, optional Dead Reckoning or GLONASS support. Among the computer’s multitude of interface connections are a VGA/DP Output, 1 LAN, 2 x RS-232, 1 x RS-422/485, 3 x DI, 3 x DO, and 3 x USB.


Gen7 Graphics proces

Learn More

Contact Featured Alliance Member:

Solutions in this blog:

Related topics:

Nexcom is an Associate Member of the Intel® Intelligent Systems Alliance


Murray Slovick

Roving Reporter (Intel Contractor), Intel® Intelligent Systems Alliance