Audio processing engine is optimized for multi-core development
AWE Core OS gives embedded audio engineers the freedom to move audio processing to the main SoC without fear of running out of CPU resources on a single core. By removing these limitations, says the company, developers and manufacturers can dramatically reduce the costs and components typically required for audio products.
“While all of our customers will eventually benefit from AWE Core OS, this is a solution directly tailored for auto manufacturers,” says Chin Beckmann, CEO and co-founder of DSP Concepts. “They blew through the limits of single core ages ago and are already pushing the limits of quad and octo-core development.”
“The increasing complexities of automotive systems are truly unmatched in consumer products,” says Beckmann. “While they are building features like fully autonomous driving, they are simultaneously innovating on the in-cabin experience: a huge component of which is audio and voice. We built AWE Core OS specifically to keep up with that innovation cycle while reducing manufacturing costs today.”
For example, says the company, the Arm Cortex family of processors are suited perfectly to benefit from AWE Core OS, enabling highly configurable software stacks appropriate to both delayed and real-time applications, such as speaker playback vs. handsfree calling. The company’s Audio Weaver end-to-end embedded audio engineering software is currently deployed across more than 40 million devices worldwide and is embedded in leading automotive audio providers Panasonic, Sony, Alpine, and more.
AWE Core OS offers the following features:
- Supports designs requiring multiple threads or various sample rates on the same SoC. For instance, a low latency telephony process can run concurrently with a normal latency entertainment post processing process.
- Enables automotive manufacturers to create more intensive and feature-rich designs that might normally exceed the capabilities of a single processing core.
- Reduces the amount of embedded software an integrator must create through Integrated TCP/IP Tuning Interface and Thread Management.
- Distributes audio processing load across multiple SMP cores.
- Provides file logging for API interactions and human readable strings on each error code for easy logging/debugging (e.g., printf).
In effect, says the company, AWE Core OS provides system designers the freedom and security in designing on a single device in a single environment.
“For example, an auto manufacturer could design a large channel count speaker playback system on one core using a large block size for code efficiency,” says Paul Beckmann, CTO and co-founder of DSP Concepts. “At the same time, using AWE Core OS, they could create a voice input and voice communication signal chain on another processor and thread using a smaller block size to achieve lower latency and perhaps a different sample rate. That’s just one small example, but it represents an improvement over today’s existing solutions.”
In February, the company announced $14.5 million in Series B funding for Audio Weaver – the result, says the company, of Audio Weaver’s success in consumer and automotive deployments in partnership with top-tier brands including Porsche, Tesla, GoPro, Braun Audio, and voice recognition partners such as Alexa Voice Service, Google Voice Assistant, as well as non-cloud connected voice recognition engines like Sensory. The company also works with several chip partners including STMicroelectronics, MediaTek, Qualcomm, Texas Instruments, NXP, and others.
If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :
