What software extracts sounds by humming them?
April 1, 2011 8:16 PM Subscribe
What's the software or academic paper that isolates sounds by humming them? I remember seeing a demonstration video of this where the narrator removed a phone ringing by imitating its trill. He also removed the guitar from the AC/DC song Back In Black. I think the title was "Isolating Sounds Via Frequency Matching" but my google-fu has failed.
Some more terms: antiphase, interference, and phase cancellation. These techniques are also used in active noise reduction. Chances are that someone a lot smarter than me will come along with a really good explanation -- in the meantime I hope this helps.
posted by safetyfork at 9:06 PM on April 1, 2011
posted by safetyfork at 9:06 PM on April 1, 2011
Best answer: Is this the paper?
Smaragdis, P.; Mysore, G.J.; , "Separation by “humming”: User-guided sound extraction from monophonic mixtures," Applications of Signal Processing to Audio and Acoustics, 2009. WASPAA '09. IEEE Workshop on , vol., no., pp.69-72, 18-21 Oct. 2009
doi: 10.1109/ASPAA.2009.5346542
Abstract: In this paper we present a novel approach for isolating and removing sounds from dense monophonic mixtures. The approach is user-based, and requires the presentation of a guide sound that mimics the desired target the user wishes to extract. The guide sound can be simply produced from a user by vocalizing or otherwise replicating the target sound marked for separation. Using that guide as a prior in a statistical sound mixtures model, we propose a methodology that allows us to efficiently extract complex structured sounds from dense mixtures.
posted by needled at 9:08 PM on April 1, 2011
Smaragdis, P.; Mysore, G.J.; , "Separation by “humming”: User-guided sound extraction from monophonic mixtures," Applications of Signal Processing to Audio and Acoustics, 2009. WASPAA '09. IEEE Workshop on , vol., no., pp.69-72, 18-21 Oct. 2009
doi: 10.1109/ASPAA.2009.5346542
Abstract: In this paper we present a novel approach for isolating and removing sounds from dense monophonic mixtures. The approach is user-based, and requires the presentation of a guide sound that mimics the desired target the user wishes to extract. The guide sound can be simply produced from a user by vocalizing or otherwise replicating the target sound marked for separation. Using that guide as a prior in a statistical sound mixtures model, we propose a methodology that allows us to efficiently extract complex structured sounds from dense mixtures.
posted by needled at 9:08 PM on April 1, 2011
Response by poster: I'm almost certain that's the paper, needled. Now if only I could find the AC/DC demo video...
posted by Monochrome at 9:29 PM on April 1, 2011
posted by Monochrome at 9:29 PM on April 1, 2011
Best answer: Here it is [mp4], at 1:05
From the author's site.
posted by qxntpqbbbqxl at 9:45 PM on April 1, 2011 [5 favorites]
From the author's site.
posted by qxntpqbbbqxl at 9:45 PM on April 1, 2011 [5 favorites]
This thread is closed to new comments.
posted by safetyfork at 8:55 PM on April 1, 2011