Voice Design API
Voice design API: create a new voice from a text description
Describe the voice you want, such as age, tone and accent, and get speech in a voice that didn't exist before.
Cloud API on the Starter and Premium plans, or the Desktop API on your own computer.
# 1. Describe the voice and hear a samplecurl -X POST https://api.vcp.name/v1/voices/design \ -H "Authorization: Bearer $VCP_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "description": "A calm older narrator with a warm, low voice", "text": "The village had been quiet for a hundred years." }' # 2. Keep the one you likecurl -X POST https://api.vcp.name/v1/voices/design/JOB_ID/save \ -H "Authorization: Bearer $VCP_API_KEY" \ -H "Content-Type: application/json" \ -d '{"name": "Old narrator"}'What developers build with it
Each call creates a fresh voice from the description, so results vary. Save the one you like and generate from it again and again.
- Casting dozens of distinct characters for a game or audio drama without recording anyone
- Brand voices that belong to no real person
- Quick voice options for users to pick from inside your app
{ "id": "04bd0571-e2c8-4d5b-a3cc-e56f85773abf", "type": "voice_design", "status": "succeeded", "result": { "audio_url": "https://files.example.com/speech.wav", "download_url": "https://files.example.com/speech.wav?download=speech.wav", "expires_at": "2026-10-31T00:18:01Z" }, "tokens_charged": 42, "finished_at": "2026-10-30T00:18:01Z"}Give your users voice design through one API key.
Endpoints
Voice design endpoints
| Method | Path | What it does |
|---|---|---|
| POST | /v1/voices/design | Design a voice and speak a sample |
| POST | /v1/voices/design/{id}/save | Save a designed voice |
| POST | /v1/speech | Speak text in a saved voice |
| GET | /v1/jobs/{id} | Get a job, or wait for it to finish |
Desktop API
Run the API on your own computer
Every desktop plan includes a local API on Windows and Mac. It needs no key and answers requests from your computer while the app is open.
curl -X POST "http://127.0.0.1:8100/api/v1/generation/design?sync=true" \ -H "Content-Type: application/json" \ -d '{ "instruct_text": "A calm older narrator with a warm, low voice", "target_text": "The village had been quiet for a hundred years.", "language": "English" }'Desktop API endpoints
| Method | Path | What it does |
|---|---|---|
| POST | /api/v1/generation/design | Generate speech in a voice made from a description |
| POST | /api/v1/generation/design/saved | Generate from a saved designed voice |
| POST | /api/v1/generation/design/saved/stream | Stream speech from a saved designed voice |
Create a key and send your first request in a few minutes.
FAQ
Common Questions
Voice design creates a new voice from a written description, with no audio needed. Voice cloning copies an existing voice from a short clip.
The model creates a new voice on every call. To keep one, save it: POST /v1/voices/design/{id}/save in the Cloud API within 24 hours, or /api/v1/generation/design/saved in the Desktop API.
In the Desktop API, designed voices can speak 600+ languages. In the Cloud API, GET /v1/models lists the languages of each model that can design voices. Set language on the request to the language of your text.
No. Voice design needs only a written description, such as age, tone, accent and pace: description in the Cloud API, instruct_text in the Desktop API. Use voice cloning to copy a real person's voice instead.
Explore
Related APIs
Text to speech
Turn text into speech with preset or cloned voices.
Voice cloning
Clone a voice from a short clip and speak new text with it.
Speech to text
Transcribe audio with word timestamps and speaker labels.
Dubbing
Dub videos into 21 languages, keeping each speaker's voice.
Subtitles
Generate SRT caption files with word-level timing.
Clips
Cut long videos into short, captioned clips.
AI Cut
Remove silences, filler words and bad takes, then render.
Avatars
Make a lip-synced talking-head video from one photo.
Voice changer
Re-voice a recording while keeping its timing and emotion.
Audiobooks
Narrate a manuscript with a voice for every character.