Forced Alignment
ElevenLabs unveils Cue for character-level audio timing
ElevenLabs introduced Cue, a forced-alignment model that timestamps characters in audio, supports more than 90 languages and aligns an hour in under a second.