Suno launches public Speech beta for spoken tracks

Suno's Speech beta turns prompts or scripts into spoken tracks of up to about eight minutes, with optional background music.

Suno launched Speech, a public beta on its web and mobile apps that turns a scene prompt or pasted script into a spoken voice, with controls for voice gender, style, and variation. The Verge says tracks last up to about eight minutes and that background music can be turned off for clean speech, while The Decoder describes spoken audio and matching background music generated as a single track. Chief product officer Jack Brody said music remains Suno's focus and that a small group tested the feature for a month. Both reports say accents and pauses are still unreliable; The Decoder specifically notes British accents sometimes sounding Australian. Cited uses include poems, meditations, and bedtime stories, in a text-to-speech market that already includes ElevenLabs, Adobe, and earlier DeepMind work. Suno has not disclosed its training data, amid copyright lawsuits from major labels and a Munich court ruling that rejected fair use.

• On 2026-10-02, Suno launched Speech, a public beta on its web and mobile apps that generates spoken audio from a scene prompt or pasted script.
• Users can set voice gender, style, and variation; The Verge says clips last up to about eight minutes and that background music can be toggled off.
• The Decoder describes Speech as producing a spoken voice and matching background music as one track, for uses such as poems, meditations, and bedtime stories.
• Chief product officer Jack Brody said music remains Suno's focus and that a small group tested Speech for a month.
• Suno warns accents and dramatic pauses are unreliable; The Decoder cites British accents sometimes sounding Australian.
• The launch enters a text-to-speech market that already includes ElevenLabs, Adobe, and earlier DeepMind work.
• Suno has not disclosed training data; major labels have sued, and a Munich court rejected fair use.