AudiCeive : mobile application to recognize unique audio frequencies using acoustic fingerprinting

Imagine asking a kid to read up a book on landmarks present in a specific country and the history of those places. How many of them would be interested to know more about those places at such a young age, let alone finish reading the book which does not earn their attention? However if we place a ph...

Full description

Saved in:
Bibliographic Details
Main Author: Saraswathi Karuppiah
Other Authors: Owen Noel Newton Fernando
Format: Final Year Project
Language:English
Published: 2017
Subjects:
Online Access:http://hdl.handle.net/10356/72803
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:Imagine asking a kid to read up a book on landmarks present in a specific country and the history of those places. How many of them would be interested to know more about those places at such a young age, let alone finish reading the book which does not earn their attention? However if we place a phone in their hand together with a movie clip playing and let them see the phone returning a few details about the place they just saw on screen, things take a different angle. The kids would be interested to continue watching the video to keep experiencing the excitement of scene information automatically updating upon change of movie scene, and in that sake, we could achieve our objective of them learning more about the places and increasing their knowledge base. This project will thus explore the usage of audio fingerprinting for educational and commercial purposes. The above scenario can be achieved by extracting the fingerprint of the real-time recorded audio and trying to match it with the already stored fingerprints of the audio portions (where scene change occurs) in the database. The details of the scenes will be stored in a separate database. Ultrasounds will be experimented with as well, since they have the advantage of being inaudible to the human ear and thus could greatly increase the accuracy of detection. For this reason, the outcome of applying the Shazam fingerprinting algorithm onto a couple of externally generated single sine tones, having frequencies above 20,000hz will be reflected mainly in this paper.