Oriol Nieto
Oriol Nieto
The new Variable Markov Oracle algorithm (included in this [PR](https://github.com/urinieto/msaf/pull/59)) didn't pass the unit tests because of two reasons: * The results of the main vmo [call](https://github.com/urinieto/msaf/blob/0.1.5-dev/msaf/algorithms/vmo/segmenter.py#L30-L32) did not follow...
In beat-sync mode (default) ground-truth boundaries need to align to closest beat-time such that labels algorithms can later use these boundaries. However, this might produce misalignments that can be confusing...
If one boundary algorithm parameter has the same name as a labeling one, the labeling one overwrites the boundary one. This shouldn't happen.
Let's do it like the Features module, and use Mixins to differentiate between the type of algorithms.
Right now the Constrained Clustering and the Shift-Invariant Probabilistic Latent Component Analysis are found in the [msaf-gpl](https://github.com/urinieto/msaf-gpl) repo due to license issues. It would be fun to re-implement them in...
Is there any way I could add an observation without the fields `time` and `duration`, such that it becomes a track-level observation (e.g., track-level tag)?
From [this](https://github.com/marl/jams/issues/196) issue I realized that the current jams schema doesn't allow for a given annotation to be curated by more than one person. Is that true? If so, we...
## 🐛 Bug I am trying to reproduce the results of the Language Identification task with the XLS-R model on the Voxligua107 dataset, but following the [current instructions](https://github.com/facebookresearch/fairseq/blob/main/examples/wav2vec/xlsr/README.md#lid-inference) yields several...
It seems like the keras' `build` function of the layers has changed its prototype and it now takes the `input_shape` as parameter, so [this](https://github.com/EderSantana/seya/blob/master/seya/layers/attention.py#L43) fails (see keras [documentation](https://keras.io/layers/writing-your-own-keras-layers/)). Regardless, once...
We should have a global file containing all the metadata in order to avoid having to manually add it after parsing the svl files. Thus, the parser could add it...