-
The NLLB model appears to have been trained with Modern Standard Arabic (MSA; _arb_) training data that is typical for common use, but not for religious texts (e.g., Biblica's Open New Arabic Version …
-
All of our new software submission issues have metadata that look like this:
```
Submitting Author: David Nicholson (@NickleDave )
All current maintainers: (@NickleDave )
Package Name: crowsetta…
-
I ran `runoak -i simpleobo:hp.obo diff -X simpleobo:hp-latest.obo -o diff.md -O md` and the `Classes added` section looked like this:
Classes added: 932
| Term |
----|
| Mosaic attenuation p…
-
## 2.7 Zaima, background
The primary purpose of zaima annotation is to indicate vocal inflection and intonation of the base text to
produce a melody when a complete passage is vocalized. Zaima is …
-
I was hoping to use DeepSqueak to conduct a study on rat vocalizations. I installed MatLab + Simulink as well as the computer vision toolbox, deep learning toolbox, curve fitting toolbox, and image p…
-
### Steps to reproduce
Run the sample app and click the `Different announcements` and `Same announcements` buttons multiple times.
### Expected results
Similar to how all the `Different annou…
-
1. segment audio into units and create spectrograms from audio using built-in good defaults for bioacoustics
2. process very large datasets in a parallelized way with the same function calls (uses `…
-
We should add some version of the algorithm that AVA uses to segment mouse USVs from audio
- https://github.com/pearsonlab/autoencoded-vocal-analysis/blob/fac3b53280bdac95dd9fb5ab8e276fdb90dab1e3/ava…
-
Mistral APIs support a bunch of options: https://docs.mistral.ai/api#operation/createChatCompletion
- `temperature`: float, [0..1], default 0.7
- `top_p`: float, [0..1], default 1
- `max_tokens`:…
-
Submitting Author: David Nicholson (@NickleDave )
All current maintainers: (@NickleDave )
Package Name: crowsetta
One-Line Description of Package: A Python tool to work with any format for annotati…