Challenge Submission
The test set utterances and all the relevant metadata are available for download in the following json file - https://drive.google.com/file/d/1vhjWcDk3UYrh2_bOb202f6LKj8jlNvLT/view?usp=sharing
The participants will share the synthesised files through a Google form -
https://forms.gle/rNmz8oduJeGQtKiY8
In case of any discrepancy, the participants will agree to share the model files or through API if requested by the organisers. The organisers may evaluate the model performance with new speakers.
Make sure you have saved the generated files with the appropriate name provided in the json file. The generated audios must be in valid .wav format.
Keep all the files in a single folder, and zip the file. The zip file name should contain the team name and track number. To submit for multiple tracks, refill the form and select the appropriate track and upload the appropriate track folder. Example folder name - "Team XYZ - Track2.zip"