Septiansyah, Hasbi (2015) Implementasi metode forensik suara pria menggunakan teknik voice recognize untuk analisis kemiripan suara pada media alat rekam telepon selular. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.
|
Text
5211100124-Paper-5211100124-paperpdf.pdf Download (496kB) | Preview |
|
|
Text
5211100124-Undergraduate_Thesis.pdf Download (4MB) | Preview |
|
|
Text
5211100124-Presentation-5211100124-presentationpdf.pdf Download (1MB) | Preview |
Abstract
Pada kasus-kasus tertentu, rekaman suara pembicaraan merupakan barang bukti digital yang memiliki peranan sangat penting untuk memberikan fakta-fakta yang diperlukan dalam proses penyidangan dan juga menunjukkan keterlibatan seseorang dengan kasus yang sedang di investigasi. Tahun 2009 Mabes Polri melaporkan sebanyak 278.537 orang pelaku tindak pidana, sebanyak 270.844 orang (97,2%) adalah laki-laki dan 7.683 orang lainnya (2,8%) adalah perempuan. Diperkirakan periode 2009-2014 sebanyak >85% pelaku tindak kriminalitas masih di dominasi oleh laki-laki. Bukti rekaman suara dapat menunjukkan identitas dari orang yang suaranya terekam pada barang bukti tersebut dengan cara melakukan pemeriksaan forensika audio untuk voice recognition dengan metode komparasi, yaitu membandingkan suara barang bukti dengan suara yang direkam sebagai pembanding. Teori voice recognition tersebut menganalisis statistik pitch, formant, bandwith dan spectogram. Jika hasil voice recognition menunjukkan bahwa suara percakapan yang ada di dalam barang bukti sama identik dengan suara pelaku, maka dapat disimpulkan bahwa suara percakapan yang ada di dalam rekaman barang bukti adalah suara milik pelaku sehingga alat rekam suara tersebut dapat digunakan sebagai bukti yang kuat di pengadilan. Hasil dari penerapan teknik forensika digital dalam penelitian eksperimen menggunakan bantuan aplikasi Praat adalah untuk menentukan kemiripan suara subjek dengan barang bukti berdasarkan hasil Pitch, Formant dan Spectogram. ================================================================================================================ In the specific cases, the recording of conversations is a digital evidence which has a very important role to provide the necessary facts in the trial process and also indicate the involvement of a person within a case that is being investigated. In 2009 as many as 278 537 the perpetrators of criminal acts which reported by Police Headquarters, which there are 270 844 people (97.2%) were male and 7683 others (2.8%) were women. It is estimated that 2009-2014 period, >85% the perpetrators of crime would still dominated by men. Evidence of the sound recording can show the identity of the person whose voice is recorded on the evidence by carrying out of checks audio forensics for voice recognition by comparative method, which compares the sound of the evidence (unknown samples) with a recorded sound as a comparison (known samples). The voice recognition theory statistical analyzes pitch, formant, bandwidth and spectogram. If the voice recognition results indicate that the voice conversation in evidence similar with voice actors, it can be concluded that the voice conversation in the recording of evidence is the voice of the perpetrator so that the voice recorder can be used as strong evidence in court , The result of applying the application of digital forensics techniques in research experiments which use Praat application assistance is for determine which actors has the resemblance voice with the evidence based on the result of Pitch, Formant, and Spectogram.
| Item Type: | Thesis (Undergraduate) |
|---|---|
| Additional Information: | RSSI 006.454 Sep i |
| Uncontrolled Keywords: | Forensika digital, audio forensic, voice recognition, suara pria |
| Subjects: | Q Science > QA Mathematics > QA76 Computer software |
| Divisions: | Faculty of Information Technology > Information System > 57201-(S1) Undergraduate Thesis |
| Depositing User: | - Taufiq Rahmanu |
| Date Deposited: | 05 Dec 2019 04:53 |
| Last Modified: | 05 Dec 2019 04:53 |
| URI: | https://repository.its.ac.id/id/eprint/72218 |
Actions (login required)
 |
View Item |
