Related Experiment Video
Updated: Aug 5, 2026
09:36
Continuous-Wave Propagation Channel-Sounding Measurement System - Testing, Verification, and Measurements
Published on: June 25, 2021
A large-scale Wi-Fi channel state information dataset for contactless human speech recognition
Rami Alazrai1,2, Hashem F Qaryouti1, Mahmoud Al-Sarayreh1
1Department of Computer Engineering, School of Computing, German Jordanian University, Amman, 11180, Jordan.
Data in Brief
|July 28, 2026
Summary
This study introduces a new dataset for Wi-Fi-based spoken-word recognition, enabling contactless speech sensing without microphones. The data captures wireless signal variations caused by speech movements for privacy-preserving applications.
Area of Science:
- Human-Computer Interaction
- Wireless Communication
- Signal Processing
Background:
- Wi-Fi sensing analyzes wireless signal variations for contactless human activity recognition.
- Wi-Fi-based spoken-word recognition offers a microphone-less alternative for speech data capture.
- Speech-associated physical movements, not acoustic waves, perturb Wi-Fi signals (CSI/RSSI).
Purpose of the Study:
- To present a novel dataset for Wi-Fi-based human speech recognition.
- To facilitate research in contactless speech sensing and privacy-preserving HCI.
- To support the development and evaluation of machine learning models for speech recognition using Wi-Fi signals.
Main Methods:
- Collected recordings of 30 English utterances from 25 subjects (22,500 trials total).
- Utilized the CSI tool to capture Wi-Fi signals from a commercial access point to a PC with an Intel 5300 NIC.
- Recorded both Channel State Information (CSI) and Received Signal Strength Indicator (RSSI) values.
Main Results:
- Developed a comprehensive dataset for Wi-Fi-based speech recognition.
- The dataset includes diverse spoken utterances and multiple trials per subject.
- Collected both CSI and RSSI data for robust signal analysis.
Conclusions:
- The presented dataset is a valuable resource for advancing Wi-Fi-based speech recognition research.
- Enables further investigation into contactless speech sensing technologies.
- Contributes to the growing field of Wi-Fi-based human sensing and its applications.
Related Concept Videos
01:03IR Frequency Region: Fingerprint Region
IR spectra are divided into two main regions: the diagnostic region and the fingerprint region. The diagnostic region of the spectrum lies above 1500 cm−1. The absorptions resulting from single-bond vibrations of the N–H, C–H, and O–H stretch at higher wavenumbers and appear on the left side of the spectrum. The stretching absorptions of the C≡C and C≡N occur between 2100–2300 cm−1. In contrast, those arising from stretching absorptions of the C=O, C=N, and C=C occur between 1600–1850 cm−1.
The...
The...
01:15Receiver Operating Characteristic Plot
A ROC (Receiver Operating Characteristic) plot is a graphical tool used to assess the performance of a binary classification model by illustrating the trade-off between sensitivity (true positive rate) and specificity (false positive rate). By plotting sensitivity against 1 - specificity across various threshold settings, the ROC curve shows how well the model distinguishes between classes, with a curve closer to the top-left corner indicating a more accurate model. The area under the ROC curve...
01:24IR Frequency Region: X–H Stretching
In IR spectroscopy, signals produced by the X−H bonds (such as C−H, O−H, or N−H) can be observed in the frequency range of 2700–4000 cm–1. The C−H stretching vibration forms sharp bands in the region 2850–3000 cm–1. The presence of the O−H stretching vibration leads to the forming of an absorption band in the frequency range 3650–3200 cm−1. At the same time, N−H stretching can be confirmed by absorption bands in the 3500–3100 cm−1 range. Even though both O−H and N−H bonds vibrate at a similar...