We proudly present the Helsinki Speech Challenge 2024 (HSC2024).

We invite researchers and scientists to test their speech enhancement and audio deconvolution algorithms on our real world data.

Participants will be provided with real-world data sets, consisting of paired clean and corrupted audio samples affected by our recording setup. The challenge is to develop innovative methods to effectively recover a clean audio signal.

The detailed description of the data sets and the challenge in general can be found from this document: https://arxiv.org/abs/2406.04123.

Audio data spectrum example.

The winners will be invited to present their results at this years Inverse Days, December 10. – 13. 2024 in Oulu, Finland.

To participate in the challenge please register using the electronic form on the rules page.