On July 1, 2024, WavLab announced XEUS, a self-supervised speech encoder trained on more than 1 million hours of audio and covering more than 4,000 languages.
On July 1, 2024, WavLab announced XEUS, a self-supervised speech encoder trained on more than 1 million hours of audio and covering more than 4,000 languages. According to the announcement, code, checkpoints, and data are being made available. Engineers can evaluate the multilingual encoder and associated data in their own systems.
To check the announcement, consult WavLab’s original publication and verify its stated scope, the availability of code, checkpoints, and data, and the training information. If you use AI to study or apply the material, avoid submitting audio or documents containing personal data without authorization, and follow your organization’s data-protection policies.