Suno has opened a beta of Speech, a model that produces spoken word and an original musical backing in one generation, the company announced on its blog on 1 October. Suno says it is the first audio model to create voice and music together as a single track rather than layering one over the other.
Speech sits inside the existing Suno app on web and mobile. A user types an idea, a poem or their own text, then describes the voice they want and the style of music behind it. The company says a small group tested it for about a month before it was opened to everyone.
Suno pitches the launch as part of what it calls creative entertainment, and says music remains at the centre of the company. It suggests meditations, pep talks, bedtime stories and overblown readings of friends' messages, a natural extension, it argues, of the birthday and wedding songs people already make on Suno. It warns the beta is rough, with accents that can drift and pauses that run long, and says it will keep changing the model as it sees how people use it.
Sources · 1