A Large Scale Audio Fingerprinting System
نویسندگان
چکیده
This paper presents a large scale audio fingerprinting system, which can be used to retrieve intended song from more than a million audio tracks. A time offset of several milliseconds could cause a number of audio fingerprints disappear. To decrease this effect, the query clip is converted to several copies with different start times. The audio fingerprints are mapped to hashes. Based on the static of the hashes on 25,000 audio tracks, it is found the frequency of occurrence does not follow the uniform distribution. A compact index structure is used to retain the quick access and none memory waste. Our experiments demonstrate the performance of this large scale audio fingerprinting system, and all of the query audio clips are captured from practical environments by various mobile devices. An iOS application is proposed based on this system and its index dataset contains more than 1 million songs.
منابع مشابه
Panako - A Scalable Acoustic Fingerprinting System Handling Time-Scale and Pitch Modification
This paper presents a scalable granular acoustic fingerprinting system. An acoustic fingerprinting system uses condensed representation of audio signals, acoustic fingerprints, to identify short audio fragments in large audio databases. A robust fingerprinting system generates similar fingerprints for perceptually similar audio signals. The system presented here is designed to handle time-scale...
متن کاملA local fingerprinting approach for audio copy detection
This study proposes an audio copy detection system that is robust to various attacks. These include the severe pitch shift and tempo change attacks which existing systems fail to detect. First, we propose a novel two dimensional representation for audio signals called the time-chroma image. This image is based on a modification of the concept of chroma in the music literature and is shown to ac...
متن کاملDaubechies Wavelets Based Robust Audio Fingerprinting for Content-Based Audio Retrieval
Audio fingerprinting is the process to obtain a compact content-based signature that summarizes the essence of an audio clip. In general, existing audio fingerprinting schemes based on wavelet transforms are not robust against large linear speed changes. The authors present a novel framework for contentbased audio retrieval based on the audio fingerprinting scheme that is robust against large l...
متن کاملContent Fingerprinting Using Wavelets
In this paper, we introduce Waveprint, a novel method for audio identification. Waveprint uses a combination of computer-vision techniques and large-scale-data-stream processing algorithms to create compact fingerprints of audio data that can be efficiently matched. The resulting system has excellent identification capabilities for small snippets of audio that have been degraded in a variety of...
متن کاملSystem Analysis and Performance Tuning for Broadcast Audio Fingerprinting
An audio fingerprint is a content-based compact signature that summarizes an audio recording. Audio Fingerprinting technologies have recently attracted attention since they allow the monitoring of audio independently of its format and without the need of meta-data or watermark embedding. These technologies need to face channel robustness as well as system accuracy and scalability to succeed on ...
متن کامل