DOI QR코드

DOI QR Code

A Study on Forgery Techniques of Smartphone Voice Recording File Structure and Metadata

스마트폰 음성녹음 파일 구조 및 메타데이터의 위변조 기법에 관한 연구

  • 박재완 (숭실대학교 글로벌미디어학부) ;
  • 곽원준 (숭실대학교 경영학부, 가톨릭대학교 인공지능학과) ;
  • 이상현 (숭실대학교 국제법무학과)
  • Received : 2022.09.30
  • Accepted : 2022.11.01
  • Published : 2022.11.30

Abstract

Recently, as the number of voice recording files submitted as court evidence increases, the number of cases claiming forgery is also increasing. If the audio recording file structure and metadata, which are objective grounds, are completely forged, it is actually impossible to detect forgery of the sophisticated audio recording file. It is extremely rare for the court to reject the file structure and metadata analysis performed with the forged audio recording file. The purpose of this study is to prove that forgery of voice recording file structure and metadata is easily possible. To this end, in this study, it was introduced that forgery detection is impossible when the 'mixed paste' function, which enables sophisticated editing based on the typification of the editing method of voice recording files, is applied. Moreover, it has been proven through experiments that forgery of file structure and metadata is possible. Therefore, a stricter standard for judging the admissibility of evidence is required when the audio recording file is adopted as digital evidence. This study will not only contribute to the standard of integrity in the adoption of digital evidence by judges, but will also contribute to the method of constructing a dataset for artificial intelligence in detecting forgery of recorded files that is expected to be developed in the future.

최근 음성녹음 파일도 법정 증거로 제출되는 수가 늘어남에 따라 위변조를 주장하는 사례도 증가하고 있다. 객관적 근거인 음성녹음 파일 구조 및 메타데이터를 완벽하게 위변조 할 경우에는 정교한 음성녹음 파일의 위변조 검출은 사실상 불가능하다. 위변조된 음성녹음 파일을 가지고 수행된 파일 구조 및 메타데이터 분석이 법정에서 거부되는 것은 쉽지 않다. 본 연구는 음성녹음 파일 구조 및 메타데이터의 위변조가 손쉽게 가능하다는 것을 증명하는 것을 목적으로 한다. 이를 위해 본 연구에서는 음성녹음 파일의 편집 방법의 유형화를 기반으로 정교한 편집이 가능한 '혼합붙여넣기' 기능을 적용할 경우 위변조 검출의 불가능함을 소개했다. 더욱이 실험을 통해 파일 구조 및 메타데이터의 위변조가 가능하다는 것을 증명했다. 따라서 음성녹음 파일이 디지털 증거로 채택됨에 있어서 더 엄격한 증거능력 판단 기준이 필요하다. 본 연구는 법관이 디지털 증거를 채택함에 무결성의 기준에 공헌할 뿐만 아니라 향후 개발될 것으로 예상되는 녹음파일 위변조 검출 인공지능을 위한 데이터셋 구축 방법에 공헌할 것이다.

Keywords

Acknowledgement

이 연구는 2020년도 숭실대학교 교내연구비 지원(융합연구)에 의한 연구임

References

  1. N.I. Park, K. Shim, and O. Jeon, "A Study on Authentication Analysis Procedure of Digital Audio Files," Journal of Digital Forensics, Vol. 13, No. 4, pp. 257-269, December 2019. DOI: 10.13064/KSSS.2022.14.3.103
  2. I. Yang, K. Kim, M. Kim, R. Baek, H. Heo, and H. Yu, "An Automatic Method of Detecting Audio Signal Tampering in Forensic Phonetics," Phonetics and Speech Sciences, Vol. 6, No. 2, pp. 21-28, June 2014. DOI: 10.13064/KSSS. 2014.6.2.021
  3. K. Lee, "Develops Smartphone Recording File Forgery Identification Technology," ChosunBiz, February 2021, https://biz.chosun.com/site/data/html_dir/2021/02/17/2021021700406.html
  4. H. Heo, B. So, I. Yang, H. Yu, "A Speech Waveform Forgery Detection Algorithm Based on Frequency Distribution Analysis," Phonetics and Speech Sciences, Vol. 7, No. 4, pp. 35-40, December 2015. DOI:10.13064/KSSS.2015.7.4.035
  5. J. Seok, "Microphone Type Classification for Digital Audio Forgery Detection," Journal of Korea Multimedia Society, Vol. 18, No. 3, pp. 323-329, March 2015. DOI: 10.9717/kmms.2015.18.3.323
  6. Supreme Court, Sentencing, 2012. 9. 13., Judgment, 2012Do 7461.
  7. O. Kwon, "A rule of thumb and it's limit at the Principle of free evaluation of evidence," Law Review Vol. 48, pp. 231-254, November 2012.
  8. Supreme Court, Sentencing, 2008. 12. 24., Judgment, 2008Do 9414.
  9. Supreme Court, Sentencing, 2015. 1. 22., Judgment, 2014Do 10978.
  10. S.J. Park and J.W. Yoon, "ENF based Detection of Forgery and Falsification of Digital Files due to Quadratic Interpolation," Journal of KIISE, Vol. 45, No. 3, pp. 311-320, March 2018. DOI: 10.5626/JOK.2018.45.3.311
  11. K. Kim, "A Study on the Forensic Application of Smartphone Recording Database," Journal of Digital Forensics, Vol. 15, No. 1, pp. 26-42, March 2021. DOI: 10.22798/kdfs.2021.15.1.26
  12. Y. Song and G. Kim, "A Study on the Detection of Falsification of Voice Recor ding Files in an Application," Journal of Digital Forensics, Vol. 16, No. 3, pp. 65-76, September 2022. DOI:10.22798/kdfs.2022.16.3.65
  13. N.I. Park, J. W. Lee, J. Kim, J. S. Lim, G. Na, and O. Jeon, "Forensic Analysis Method for Forgery Detection of Call Recordings Generated by Samsung Smartphones," Journal of Digital Forensics, Vol. 16, No. 1, pp. 142-159, March 2022. DOI:10.22798/kdfs.2022.16.1.142
  14. E. Jang and J. Shin, "A Proposal on Data Modification Detection System using SHA-256 in Digital Forensics," The Journal of the Institute of Internet, Broadcasting and Communi cation, Vol. 21, No. 4, pp. 9-13, August 2021. DOI:10.7236/JIIBC.2021.21.4.9
  15. S. Lee and J.W.Park, "Research on Improving Standards of Admissibility of Digital VoiceRecorded File under Criminal Procedure Law -Centering on the Possibility of Case Law Emphasizing Hash Value -," Yonsei Law Review, Vol. 32, No. 3, pp. 31-59, September 2022. DOI:10.21717/ylr.32.3.2
  16. S.H. Moon, "Analysis of AI-Applied Industry and Development Direction," The Journal of the Convergence on Culture Technology, Vol. 5, No. 1, pp. 77-82, February 2019. DOI: 10.17703/JCCT.2019.5.1.77