CJ Carr

Hope it gets implemented!

> @PeterL1n Since this Op is very similar to ScatterNd op, we should be able to add the support fairly soon. [TensorScatterUpdate](https://www.tensorflow.org/api_docs/python/tf/raw_ops/TensorScatterUpdate) from the tensorflow docs: "This operation is very...

I am also curious about this

We successfully fine-tuned the 1b model in the cloud on p3.8xlarge, on a custom dataset, over a few days.

this is definitely helpful, and narrows down the genre search however, this is not the list of artist/genre labels from the training data. for example, the text file lists `tony...

Thanks! Would spherical interpolation be a better choice?

Hi Stefan, Replicating other loudness standards would be useful. I'm okay with A-weighting. I'm looking to get validation that - The above is an appropriate and efficient way to arrive...

@bmcfee Not quite. I'm looking for 1D loudness and a fast/precise method of its calculation. It seems it's better to apply perceptual weighting to the power spectrogram _before_ calculating the...

> If you want frame-wise loudness, you can compute log_S as above and then sum over the rows: loudness = log_S.sum(axis=0, keepdims=True). I thought decibels don't sum like that. You...

you’re doing the god’s work On Wed, Jan 11, 2023 at 10:29 PM itsnotlupus ***@***.***> wrote: > Things seem to be working as described above. Here's what a ~6 second...