Development of an acoustically adaptive modular system for near real-time clarity-enhancement

Alexander Liu Cheng*, Patricio Cruz, Nestor Llorca Vega, Andrés Mena

*Corresponding author for this work

Research output: Chapter in Book/Conference proceedings/Edited volumeConference contributionScientificpeer-review

55 Downloads (Pure)


This paper details the development of an acoustically adaptive modular system capable of enhancing Speech Clarity (C50 Clarity Index) in specific locations within a space in near real-time. The mechanical component of the system consists of quadrilateral, truncated pyramidal modules that extend or retract perpendicularly to their base. This enables said modules (1) to change in the steepness of the sides of their frustum, which changes the way incoming sound waves are deflected/reflected/diffused by the surfaces of the pyramid; and (2) to reveal or to hide the absorbent material under each module, which enables a portion of incoming sound waves to be absorbed/dissipated in a controlled manner. The present setup considers a fragmentary implementation of six modules. The behavior of these modules is determined by two steps in the computational component of the system. First, the initial position of the modules is set via a model previously generated by an evolutionary solver, which identifies the optimal extension/retraction extent of each of the six modules to select for individual configurations that collectively ascertain the highest clarity in said specific locations. Second, a simulated receiver at the location in question measures the actual clarity attained and updates the model’s database with respect to the configuration’s corresponding clarity-value. Since the nature of acoustics is not exact, if the attained measurement is lower than the model’s prediction for said location under the best module-configuration, but higher than the second-best configuration for the same location, the modules remain at the initial configuration. However, if the attained values are lower, this step reconfigures the modules to instantiate the second—or third-, fourth-, etc.—best configuration and updates the model’s database with respect to the new optimal module-configuration value. These steps repeat each time the user moves to another specific location. The objective of the system is to contribute to the intelligent and intuitive Speech Clarity regulation of an inhabited space. This contributes to its Interior Environmental Quality, which promotes well-being and quality of life.

Original languageEnglish
Title of host publicationAmbient Intelligence
Subtitle of host publicationProceedings of the 15th European Conference (AmI 2019)
EditorsIoannis Chatzigiannakis, Boris De Ruyter, Irene Mavrommati
Number of pages16
ISBN (Electronic)978-3-030-34255-5
ISBN (Print)978-3-030-34254-8
Publication statusPublished - 2019
EventAml 2019: 15th European Conference on Ambient Intelligence - Rome, Italy
Duration: 13 Nov 201915 Nov 2019

Publication series

NameLecture Notes in Computer Science
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


ConferenceAml 2019: 15th European Conference on Ambient Intelligence

Bibliographical note

Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project

Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.


  • Adaptive acoustics
  • Cyber-physical systems
  • Internet of Things


Dive into the research topics of 'Development of an acoustically adaptive modular system for near real-time clarity-enhancement'. Together they form a unique fingerprint.

Cite this