Dolphin Design, a semiconductor designer of power management, audio, and processing devices with in-house ASIC design services, announced the launch of WhisperExtractor, a mixed-signal technology for voice and audio-enabled SoCs, enabling voice and sound classification at a record µW level power budget. The new technology paves the way to breakthrough always-on voice user interface, real time speech-to-text, and always-on sound detection.
WhisperExtractor is a disruptive technology that addresses a major challenge in the voice-user interaction space - low power consumption. With the increasing adoption of voice user interfaces in everyday objects, it has become critical to have efficient and low-power solutions for voice processing.
According to the French company, this new IP enhances battery-powered applications aiming at Keyword Spotting (KWS), Automatic Speech Recognition (ASR), Natural Language Processing (NLP), or power-efficient sound classification such as intelligent security cameras. The new technology uses a mixed-signal architecture to efficiently extract the Mel Frequency Cepstrum Coefficients (MFCC) needed for those applications.
By doing so, it reduces power consumption by up to 99%, enabling always-on listening at a power budget of just µW, on a 32kHz clock. This disruptive technology allows for constant monitoring of the voice without the need for frequent charging or compromising battery life, and is an excellent solution for any low-power, always-on, voice interface SoC.
How does WhisperExtractor works?
Conventional signal processing methods to process Mel Frequency Spectrum Coefficients (MFCC) require a significant amount of energy to convert analog signals to digital and perform signal processing calculations and advanced software capabilities. It makes it difficult to implement always-on voice listening or sound classification without draining too much current from the battery.
WhisperExtractor is based on mixed-signal architecture to efficiently extract the MFCC needed for audio classification without requiring any software programming. It connects directly to an analog microphone and outputs the data needed by an AI accelerator block to handle inference/classification. This new approach shatters the limitations of conventional practices, empowering power consumption beyond previous boundaries. The advanced architecture of this innovative IP also frees massive headroom on DSP or CPUs and wakes up the system only when necessary.
"We are proud to bring this cutting-edge technology to the IP market," says Hakim Jaafar VP Marketing at Dolphin Design. "WishperExtractor furthers our investment in our comprehensive set of IPs that enable the cost-effective design of power-efficient SoCs. Be it our audio converters, power management, or processing IPs. With this ground-breaking IP, we prove once again that innovation is part of our DNA, so stay tuned!"
The Dolphin Design IP catalog encompasses more than 300 references of audio converters, power management, and processing IPs that are commodity features in SoCs. These IPs, which are proven in multiple foundries and processes enable numerous fabless and system makers that aim at designing a more energy-efficient SoC while focusing their in-house resources on more valuable tasks to differentiate their SoC.
www.dolphin-design.fr
- on Industry News
- News
Dolphin Design Unveils Innovative IP for Sound Classification Reducing Power Requirements by 99%
June 27 2023, 05:10
Dolphin Design, a semiconductor designer of power management, audio, and processing devices with in-house ASIC design services, announced the launch of WhisperExtractor, a mixed-signal technology for voice and audio-enabled SoCs, enabling voice and sound classification at a record µW level power budget. The new technology paves the way to breakthrough always-on voice user interface, real time speech-to-text, and always-on sound detection.