Dan Ellis

Centering can make a big difference when unwrapping phase along frequency, but I don’t know what reassignment actually entails. DAn. On Fri, Feb 5, 2021 at 09:36 Brian McFee wrote:...

Once you get to variable rates, you have to make sure you are clear about how the time base is defined, because now the input and output time bases have...

See for example: https://github.com/dpwe/pitchfilter/blob/master/resample_map.m (which, weirdly, has been translated into Matlab somewhere between my memory and GitHub). That code does "time scaling" by simple resampling, but otherwise it's a similar...

The constraint is bi-directional. I can't decide whether to include this peak until I see what comes after: is it a local max (ok, 1 step fixed delay), but also...

I don't get it. My intuition for DP is that I work up to time T, and at that time I can figure out whether to include or not include...

What’s the point of win_length independent of frame_length? If the frame samples outside of the win are zero’d out anyway? Unless we also allow win_length > frame_length, which I can...

On Sat, Apr 14, 2018 at 9:10 AM, Brian McFee wrote: > What’s the point of win_length independent of frame_length? If the frame > samples outside of the win are...

Also a problem for me. I'm willing to do some legwork if it helps. Trying to get my minim-based [oscilloscope](https://www.labrosa.org/~dpwe/resources/Processing/) working under current mimim.

This worked. Running the xattr change on the sound/ and minim/ directories: ``` $ xattr -r -d com.apple.quarantine ~/Documents/Processing/libraries/sound/* $ xattr -r -d com.apple.quarantine ~/Documents/Processing/libraries/minim/* ``` then launching Processing and...

You could try reading Avery Wang's papers e.g. https://www.ee.columbia.edu/~dpwe/papers/Wang03-shazam.pdf The point about the spectral-peak-pair landmarks is that some of them remain invariant to channel and additive noise, giving the representation...