The Pulse
Suno Adds Speech Model That Pairs Spoken Voice With Music
Suno opened its Speech beta to users on October 1, 2026, combining spoken voice and original background music in one generated track. The feature is available on web and mobile, with Suno warning that accents and dramatic pauses can still b
AI.info Team · Written with AI ·
“Today, we’re expanding what's possible in Suno with Speech: the first audio model that generates voice and music together as one cohesive track.”
Jack Brody, chief product officer, Suno
Suno puts narration and its soundtrack in one generation
Suno opened its Speech beta to all users on October 1, adding spoken audio backed by original music to its creation platform. The company describes Speech as a model that generates voice and music together in a single track, rather than asking users to produce narration and then add a score separately. Suno’s release notes say the feature is available on web and mobile.
The workflow starts with text: users enter an idea, a poem or other writing, then describe the voice and musical style they want. Suno says the result combines the spoken material with a newly generated backing track. Its announcement does not specify controls for editing the narration and music independently after generation.
From bedtime stories to “ASMR grocery lists”
Suno frames Speech as a way to turn ordinary writing and personal messages into short audio creations. Its examples include poems, meditations, pep talks and bedtime stories, along with dramatic readings of friends’ texts and voice notes paired with elaborate scores. The release notes offer a similar range, from bedtime stories over soft piano to hype speeches backed by stadium drums.
That puts the feature closer to narrated audio with a custom soundtrack than to Suno’s familiar prompt-to-song workflow. A user can specify both the spoken delivery and the musical setting in one request, according to the company’s description. Suno presents the combination as a new format for personal or playful creations, not as a separate professional voice-production suite.
The beta still has audible rough edges
Suno says it tested Speech with a small group of users for the past month before opening the beta more widely. The company cautions that the output remains under development: British accents may “wander off to Australia and back,” and dramatic pauses can become overly dramatic. Those examples point to limits in how consistently the model follows requested speech characteristics and pacing.
The announcement gives no benchmark results, supported-language list, pricing details or technical description of the model. Suno instead asks users to try the beta and provide feedback as it improves the feature. Its public description establishes the central product change—a single generation that joins speech and music—while leaving performance measures and finer-grained controls unspecified.
A new format inside Suno’s music platform
Speech extends Suno beyond songs by giving written material a spoken performance and a musical setting through the same creation flow. That may suit creators looking for a personalized story, message or short audio piece without assembling separate narration and soundtrack tools. Whether the output can meet more exacting needs will depend on how reliably the beta handles voices, timing and the requested musical style.
For now, the clearest caveat comes from Suno itself: accents can drift and pauses can overshoot. Speech is available to users on web and mobile, and the company says it plans to keep improving it based on what people make and where the results fall short.