Related Experiment Video
Updated: Jan 17, 2026
05:48
Author Spotlight: Investigating the Impact of Emotional Prosodies on Voice Recognition and Perception
Published on: August 9, 2024
2.0K
Analogue speech recognition based on physical computing
Mohamadreza Zolfagharinejad1, Julian Büchel2, Lorenzo Cassola1,3
1NanoElectronics Group, MESA+ Institute and BRAINS Center for Brain-Inspired Computing, University of Twente, Enschede, the Netherlands.
Nature
|September 17, 2025
Summary
We developed an efficient edge temporal-signal processor using in-materia computing. This system achieves high accuracy for speech commands with low power consumption, enabling advanced edge AI applications.
Area of Science:
- Computer Science
- Materials Science
- Electrical Engineering
Background:
- Decentralized computing in edge devices (IoT, autonomous driving, healthcare) requires efficient, low-power processing of time-dependent signals.
- Traditional processors face limitations due to the von Neumann bottleneck and domain conversions, hindering edge system performance.
- In-materia computing offers a novel approach to overcome these limitations by performing computation within the material itself.
Purpose of the Study:
- To propose and demonstrate an edge temporal-signal processor utilizing in-materia computing for efficient feature extraction and classification.
- To achieve near-software accuracy on benchmark speech datasets with significantly reduced latency and power consumption.
- To advance the development of compact, efficient, and high-performance heterogeneous smart edge processors.
Main Methods:
- Developed a nonlinear, room-temperature reconfigurable-nonlinear-processing-unit layer for analogue, time-domain feature extraction from raw audio signals.
- Implemented a compact analogue in-memory computing chip with memristive crossbar arrays for neural network classification.
- Trained and evaluated the system on the TI-46-Word and Google Speech Commands datasets.
Main Results:
- Achieved near-software accuracy for speech command recognition on benchmark datasets.
- Demonstrated submillisecond latency for the complete processing pipeline.
- Reported low energy consumption: ~300 nJ for feature extraction and ~78 µJ for classification (with potential down to ~10 µJ).
Conclusions:
- The proposed in-materia computing-based edge temporal-signal processor offers a promising solution for efficient edge AI.
- The system's high accuracy, low latency, and minimal power consumption pave the way for advanced, heterogeneous smart edge devices.
- This work highlights the potential of in-materia computing to revolutionize edge processing for time-dependent signals.
Related Concept Videos
01:06Echo
889
The human ear cannot distinguish between two sources of sound if they happen to reach within a specific time interval, typically 0.1 seconds apart. More than this, and they are perceived as separate sources.
Imagine the sound is reflected back to the ears. Assuming that the source is very close to the human, the difference between hearing the two sounds—the emitted sound and the reflected sound—may be more than the minimum time for perceiving distinct sounds. If this is the case,...
Imagine the sound is reflected back to the ears. Assuming that the source is very close to the human, the difference between hearing the two sounds—the emitted sound and the reflected sound—may be more than the minimum time for perceiving distinct sounds. If this is the case,...
889
01:21Perceiving Loudness, Pitch, and Location
942
The human brain perceives pitch through two primary mechanisms reflected in place theory and frequency theory. Each mechanism describes how sound waves are interpreted as specific pitches by the brain, offering insights into the intricate processes of auditory perception.
Place theory, or place coding, suggests that different pitches are heard because various sound waves activate specific locations along the cochlea's basilar membrane. The brain determines the pitch of a sound by...
Place theory, or place coding, suggests that different pitches are heard because various sound waves activate specific locations along the cochlea's basilar membrane. The brain determines the pitch of a sound by...
942
01:23Design Example
530
The innovation of touch-tone telephony revolutionized the telecommunications industry by replacing the traditional rotary dial with a dual-tone multi-frequency (DTMF) signaling system. This system uses a matrix-style keypad with buttons arranged in four rows and three columns, creating 12 distinct signals each assigned to a pair of frequencies. Each button press results in a simultaneous generation of two sinusoidal tones – one from a low-frequency group (697 to 941 Hz) and one from a...
530
01:01Perception of Sound Waves
5.4K
The human ear is not equally sensitive to all frequencies in the audible range. It may perceive sound waves with the same pressure but different frequencies as having different loudness. Moreover, the perception of sound waves depends on the health of an individual's ears, which decays with age. The health of one's ears may also be affected by regular exposure to loud noises.
The pitch of a sound depends on the frequency and the pressure amplitude of the source. Two sounds of the same...
The pitch of a sound depends on the frequency and the pressure amplitude of the source. Two sounds of the same...
5.4K