New benchmark for higher-quality lyrics transcription from AudioShake Research 📝
AudioShake paper sets a new direction for high-quality lyrics transcription by proposing a formatting-aware evaluation benchmark and a revised lyrics dataset.
In a year of historic AI innovation, TIME selected AudioShake as one of this year’s inventions changing “how we live, work, play, and think about what’s possible.”
AudioShake helps unblock a long-standing challenge in dubbing, transcription, and captioning technology with a new dialogue, music, and effects separation model.
AudioShake Live is a platform that lets labels, publishers, and their partners bring our A.I. technology “in-house,” and create instrumentals and stems on demand.