Today, many people have problems understanding the speech content of a movie, e.g. due to hearing impairments. This paper describes a method for improving the speech intelligibility of movie sound. Speech is detected by means of a pattern recognition method; the audio signal is then attenuated during periods where speech is absent. The speech signals are further processed by a spectral weighting method aiming at the suppression of the background noise. The spectral weights are computed by means of feature extraction and a neural network regression method. The output signal finally carries all relevant speech with reduced background noise allowing the listener to follow the plot of the movie more easily. Results of numerical evaluations and of listening tests are presented.
https://www.aes.org/e-lib/browse.cfm?elib=14779
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.
Learn more about the AES E-Library
Start a discussion about this paper!
