Narration voice

audio Queued Runs in the background — you get a link when it is ready.

Turn up to a minute of text into natural narration in a named voice.

Built for

Voiceovers for shorts
Listing narration
Explainers

How it works

You provide

The words to say (up to 1,000 characters), and optionally a voice, a delivery note and a language.

You get back

One audio file (MP3 by default) you can lay under a video.

What this means for you

Thirty named voices across 80+ languages.
Tell it how to sound — warm, brisk, documentary.
One flat price per request, up to a minute of speech.

Try it

Use the API

Details

What you can set

Setting	Type and limits	What it means
`voice` optional	text, 1–40 characters	Named narrator voice; a calm, clear default when omitted.
`prompt` required	text, 1–1000 characters	The words to say — up to 1,000 characters (about a minute of speech). Expressive tags like [sigh] or [whispering] are allowed.
`language_code` optional	text, 2–16 characters	Language of the speech (BCP-47 code such as en-US or es-ES); detected when omitted.
`output_format` optional	text, one of: mp3, wav, ogg_opus	File type of the result; mp3 when omitted.
`style_instructions` optional	text, 0–500 characters	How to deliver it, in plain words (warm, brisk, like a documentary…).

What you get back: An audio file (MP3 by default)
Limits: Max input 1000 chars
How it runs: Queued — Runs in the background — you get a link when it is ready.
Alias: oneup/voice-standard

For developers

The same request as an API call, with this capability's example input:

curl -X POST https://api.oneupai.com/v1/queue/submit/oneup/voice-standard \
  -H "Authorization: Bearer $ONEUP_API_KEY" \
  -H "Idempotency-Key: $(uuidgen)" \
  -H "Content-Type: application/json" \
  -d '{
  "prompt": "Welcome home. Three bedrooms, two baths, and a garden that gets the morning sun.",
  "style_instructions": "warm, unhurried, like a friend showing you around"
}'

Show input schema (JSON)

{
  "type": "object",
  "required": [
    "prompt"
  ],
  "properties": {
    "voice": {
      "type": "string",
      "maxLength": 40,
      "minLength": 1,
      "description": "Named narrator voice; a calm, clear default when omitted."
    },
    "prompt": {
      "type": "string",
      "maxLength": 1000,
      "minLength": 1,
      "description": "The words to say — up to 1,000 characters (about a minute of speech). Expressive tags like [sigh] or [whispering] are allowed."
    },
    "language_code": {
      "type": "string",
      "maxLength": 16,
      "minLength": 2,
      "description": "Language of the speech (BCP-47 code such as en-US or es-ES); detected when omitted."
    },
    "output_format": {
      "enum": [
        "mp3",
        "wav",
        "ogg_opus"
      ],
      "type": "string",
      "description": "File type of the result; mp3 when omitted."
    },
    "style_instructions": {
      "type": "string",
      "maxLength": 500,
      "description": "How to deliver it, in plain words (warm, brisk, like a documentary…)."
    }
  },
  "additionalProperties": false
}

API Docs

Other audio capabilities

Studio voice

Our most natural voice — the one for ad reads, trailers and anything a customer will judge you by.

18 credits

Original song

Describe the style, hand over your lyrics, and get back a finished song you own the rights to use.

6 credits

Sound effect

Any sound you can describe — a door, a crowd, rain on a roof — as a clip you can drop into an edit.

8 credits

Transcript for captions

Turn the speech in your audio or video into a word-timed transcript your captions can follow, word by word.

2 credits per started minute of your media