Submissions
Submisison are now open and due on Feburary 13th, 2023
Please make submissions directly in our open review: SLU Grand Challenge ICASSP Submission Link
Below we describe the submission format.
All tracks are required to submit a 2-page paper describing the method used to achieve the results following the ICASSP Paper Guidelines.
Additionally please share a single .zip
file containing model summary and predictions on the test set as supplemental material, descriptions for each track are listed below.
Quality Track:
quality_model_summary.txt
: Use Model Summary Formatquality_model_predictions.txt
: All model predictions on overall test set, use Model Predictions Format
On-device Track:
ondevice_model_summary.txt
: Use Model Summary Formatondevice_model_predictions.txt
: All model predictions on overall test set, use Model Predictions Format
Low Resource Track:
lr_weather_model_summary.txt
: Model summary of the model used for the weather domain, Use Model Summary Formatlr_reminder_model_summary.txt
: Model summary of the model used for the reminder domain, Use Model Summary Formatlr_weather_predictions.txt
: All model predictions on weather test set, use Model Predictions Formatlr_reminder_predictions.txt
: All model predictions on reminder test set, use Model Predictions Format
Model Summary Format: Reported EM and File Size as a text file, an indication whether the techniques used are open source, and a team name. Model Predictions Format: A TSV file containing, file_id, and model prediction as a string