-
Hello, I've trained a model for a while using the solo_instrument config at 48 kHz, but the audio is still fairly noisy even after 117k steps (spectral loss is ~9 on average).
I'd like to continue …
-
Hi, I'm trying to run the code from the [`train_autoencoder` notebook](https://github.com/magenta/ddsp/blob/main/ddsp/colab/demos/train_autoencoder.ipynb) with a newer DDSP version (3.2.0). I made a l…
-
Traceback (most recent call last):
File "train_diff.py", line 86, in
train(args, initial_global_step, model, optimizer, scheduler, vocoder, loader_train, loader_valid)
File "/home/csmxj/DD…
-
Upon closing the plugin, the Pitch/Volume XY knob resets to default and there is no way to save it. It's fine if I just close it, But opening it again is what resets it. Changing the plugin suspension…
-
**Is your feature request related to a problem? Please describe.**
I tried to put rift feedback lite after ddsp in the chain hoping ddsp would pass on the midi information of its pitch detector, so i…
-
**Describe the bug**
The colab web trainer errors when it starts training
`
Training...
2023-08-07 08:10:28.967989: W tensorflow/compiler/xla/stream_executor/platform/default/dso_loader.cc:64] C…
-
Hello,
Some time ago, I noticed [a bug in `compute_loudness` in magenta/ddsp](https://github.com/magenta/ddsp/issues/361), and I think this repo has a similar issue. Below is the loudness curve extra…
-
1. Generating tfrecords with
`ddsp_prepare_tfrecord \`
` --input_audio_filepatterns /path/to/audio/* \`
` --output_tfrecord_path /path/to/train.tfrecords \ `
` --alsologtostderr \`
`-…
iCorv updated
2 years ago
-
Hi.
I'd trained this ddsp model with my own dataset with ``ddsp.training.models.Autoencoder()``. I'd like to use [magenta-js](https://github.com/magenta/magenta-js) with my own model. This magenta-…
-
Could this be used to replace the filtered noise part in the original way of using ddsp with f0 input? potentially adding spectral centroid as an optional input too?