Suno API Speech: Turn a Script Into Narrated Audio

Q: Does the Speech API create a song?

No. It creates spoken audio, optionally with background music. Use the music generation or custom generation endpoint when you need singing.

Q: How is the speech duration controlled?

Duration follows the script length. The endpoint does not support a duration parameter and rejects one with HTTP 400.

Q: Can I use a cloned voice?

The Speech endpoint does not currently support voice cloning. Use gender to request a male or female speaker.

Generate speech and narration with Suno API: send a script, set tone, gender and backing music, then poll the song ID and download the finished audio.

SSuno-API team Oct 31, 2026 2 min read

A speech script transformed into narrated audio with background music

Photo: Pexels

Short answer: call POST /api/music/speech, put the full text in script, and describe delivery with tone. Set gender and backing_music when needed. A normal request returns about two spoken-audio candidates; store each song_id for polling and download.

Speech is not singing

Goal	Endpoint	Output
Speech, narration or voiceover	`/api/music/speech`	Spoken audio with optional music
Song with melody	`/api/music/create`	Sung music
Custom lyrics and style	`/api/music/create/custom`	Custom-mode music

Minimal request

curl -X POST "https://www.suno-api.io/api/music/speech" \
  -H "Authorization: Bearer $SUNO_API_KEY" \
  -H "Content-Type: application/json" \
  --max-time 300 \
  -d '{
    "script": "Good evening everyone. Thank you for joining us today."
  }'

A normal run currently costs 0.6 CNY and usually returns two candidates.

Control the delivery

{
  "script": "Good evening everyone. Thank you for celebrating this day with us.",
  "tone": "warm and slow, wedding reception",
  "song_title": "Wedding Toast",
  "gender": "f",
  "backing_music": false,
  "model": "suno-v6"
}

Field	Purpose	Limit / note
`script`	Full speech text or a description	Up to 5000 characters for v6
`tone`	Emotion, pace and setting	Up to 1000 characters
`gender`	Male or female speaker	`m` or `f`
`backing_music`	Music behind the narration	Defaults to true; false gives dry speech
`song_title`	Output title	Up to 100 characters

After generation

Save the returned song_id.
Poll POST /api/music/query until completed or playable.
Request media through /api/music/download-url or /api/music/download-file.
Copy important files to your own storage.

Common validation errors

Error	Cause	Fix
`script is required for speech`	No non-empty script	Send a script
`duration is not supported`	Unsupported duration field	Remove duration and use script length
Character limit error	Script, tone or title is too long	Trim to the reported count
Copyright rejection	Restricted source text	Use your own wording; failed generation is refunded