In this article, an evaluation of a recently published hum detection algorithm for audio signals is presented. To determine the performance of the method, large amounts of artificially generated and real-world audio data, containing a variety of music and speech recordings, are processed by the algorithm. By comparing the detection results with manually determined ground truth data, several error measures are computed: hit and false alarm rates, frequency deviation of the hum frequency estimation, offset of detected start and stop times and the accuracy of the SNR estimation.
Click to purchase paper as a non-member or login as an AES member. If your company or school subscribes to the E-Library then switch to the institutional version. If you are not an AES member and would like to subscribe to the E-Library then Join the AES!
This paper costs $33 for non-members and is free for AES members and E-Library subscribers.