TTS API

Text to speech API: one POST request, one mp3

REST API with Bearer keys. Send text and a voice id — get an mp3 back. 200,000 characters a month for €20, speech-to-text included.

Start free — no card → Listen to examples

Live examples

Hear it before you sign up

Recorded by the service itself, unedited. Each file carries an AI label.

TRK 01EN0:00

Text: «This audio was generated by a single POST request to the Alvenuri API. It took less than a second.»

TRK 02RU0:00

Text: «Счёт на 22 рубля 50 копеек от 3 марта 2026 года. Доставка займёт 5 дней, к оплате — 1 250 рублей.»

POST /api/v1/tts

curl https://alvenuri.com/api/v1/tts \
  -H "Authorization: Bearer vs_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{"voice_id": "supertonic-f4", "lang": "en",
       "text": "Order 1042 ships on March 3rd.", "mode": "normal"}' \
  -o speech.mp3

Simple contract

POST /api/v1/tts with text, voice_id, lang and mode. The response body is the mp3; headers mark it as AI-generated.

Queue for long jobs

POST /api/v1/tts/jobs returns a job id; poll it and download the audio when it’s done.

Voice list and STT

GET /api/v1/voices lists voices; POST /api/v1/stt turns speech back into text with word timings.

Same quality as the studio

Number normalization, stress marks and reading modes work exactly as in the browser.

How it works

Three steps from text to mp3

01
Paste text
10,000 characters a month on the free plan, up to 5,000 per run. Numbers, dates and amounts are fine as digits.
02
Pick a voice and pace
Calm, normal or fast reading, a pause between sentences, [pause 5s] tags where you need silence.
03
Download mp3
The file is ready in seconds and stays in your history. On Pro the same works over the API.

Pricing