Sound separation by humming.
Paris Smaragdis, Gautham J. Mysore · The Journal of the Acoustical Society of America · 2010
We present a novel interface for selecting and isolating sounds from complex audio mixtures. Traditional interfaces in audio editors provide graphical representations of sounds through which a user might be able to visually identify elements of sounds in a mixture; they do not, however, facilitate object-specific editing (e.g., selecting only the voice of a singer in a song). The proposed interface uses audio guidance from a user in order to specify a target sound within a mixture. The user is asked to vocalize the desired target sound, and an automated process identifies and isolates the elements of the mixture that best relate to the user’s input. Such a tool makes the extraction of complex sources an easier process than using graphical tools or statistical source separation models.