Spirometric Dataset

An open dataset of respiratory sound signals for research on spirometric parameter estimation. About The Dataset In this repository, we provide a systematically labeled dataset of audio recordings from 85 healthy subjects (43 male, 42 female), aged between 19 and 66 years. Each participant performed spirometry tests using the Contec SP-10 spirometer, while simultaneously being recorded with the Thinklabs ONE digital stethoscope. Six manoeuvres were performed per participant, with the stethoscope placed on both sides of the neck. All recordings were captured in mono at a sampling rate of 16 kHz. Each signal was processed with a median filter to suppress transient noise and heartbeats, followed by zero-padding to fixed all recordings to 6-second duration. The dataset contains 510 respiratory sound signals (six recordings per participant), each corresponding to a valid expiratory manoeuvre performed by an anonymized subject. All files are in .wav format, organized into participant-specific folders with no personally identifiable information and documented in a central Excel file that links each audio recording to its corresponding data. Each manoeuvre is uniquely identified by an anonymized participant code followed by the manoeuvre number. For each recording, the Excel file includes: biological features — gender, age, height and weight; and measured spirometric parameters — FEV1, FVC, PEF, FEV1/FVC ratio, FEF25-75%, FEF25, FEF50, FEF75 and ERV. Additionally, the file contains an 'observations' field reserved for future specialized pulmonologist annotations related to specific respiratory disorders, such as type, severity or additional clinical notes File Format To preserve anonymity, filenames contain no personal information and follow a consistent format that encodes only the recording timestamp and manoeuvre number. This resource is designed to support the development and evaluation of algorithms for estimating spirometric parameters from respiratory audio. Filenames encode only the timestamp of the recording and the associated manoeuvre number, following the structure: <YYYY><MM><DD><HH><mm>_<N>.wav where: YYYY: year (4 digits) MM: month (2 digits) DD: day (2 digits) HH: hour (24-hour format, 2 digits) mm: minutes (2 digits) N: manoeuvre number (1 digit) For example, the filename 202501231726_4.wav represents a recording from January 23, 2025, at 17:26, corresponding to the 4th expiratory manoeuvre performed by the subject. This naming convention ensures subject de-identification while preserving temporal traceability and enabling unambiguous association between each audio file and its corresponding expiratory event. License Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) Copyright (c) 2026 QHPC & SP Research Lab This dataset is licensed under the Creative Commons Attribution-NonCommercial 4.0 International License. You are free to: - Share — copy and redistribute the material in any medium or format - Adapt — remix, transform, and build upon the material Under the following terms: - Attribution — You must give appropriate credit. - NonCommercial — You may not use the material for commercial purposes. Full license text: https://creativecommons.org/licenses/by-nc/4.0/ Contact Alejandro Antonio Salvador Navarro ([e-mail])University of Jaén Francisco Jesús Cañadas Quesada ([e-mail])University of Jaén Juan de la Torre Cruz ([e-mail])University of Jaén Funding This work was supported in part by the Grant PID2023-146520OB-C21 funded by MICIU/AEI/10.13039/501100011033 and, as appropriate, by “ERDF A way of making Europe”, by “ERDF/EU”, by the “European Union” or by the “European Union NextGenerationEU/PRTR