EN
Start free
AI voice platform for teams TTS · STT · Cloning · API

Turn every script into broadcast-ready audio.

Voixa AI lets you write, voice and transcribe in Vietnamese, English and eight more languages. Work in the Studio, or put the same voices inside your product with one API call.

Free tokens every day. Pay only for what you use.

Chương 3 — Người gác đèn

Ngọn hải đăng bật đèn lúc hoàng hôn, như mọi buổi chiều suốt ba mươi năm qua. Người gác đèn vẫn ngồi đó, nhìn ra biển và chờ con tàu cuối cùng trở về. Gió bắt đầu nổi, mang theo mùi muối và tiếng sóng vỗ vào ghềnh đá phía dưới.

Ông kéo chiếc ghế gỗ lại gần cửa sổ, rót một chén trà đã nguội từ lâu.
412 / 5,000 · ≈ 25 sGenerate
NP
Nhật PhongVietnamese · Male · narrative
TD
Thùy DungFemale · warm
MT
Minh TriếtMale · documentary
LN
Lighthouse NarratorClone · your team
AL
AlbaFemale · English
Chương 2 — Rời bến
0:32 / 1:18

Studio: write, pick a voice, generate. Projects keep the team’s clips together.

Features

Everything between the script
and the published file.

Eight tools on the same voices, projects and API keys.

Studio

Write, preview and generate clips up to 5,000 characters. A searchable library, organised by project.

  • Projects
  • Library search
  • WAV download

Long-form podcasts

Up to 100,000 characters, about eighty minutes, voiced in parallel and joined into one file.

  • One WAV
  • Public episode page

Voice cloning

Your brand voice from a ten-second sample. Clones need the speaker’s consent and stay private to your account.

  • Record or upload
  • Private by default

Voice design

Describe the voice you need, such as a warm old storyteller or a calm news anchor, and pick one of three takes. Save it and it works like any of your voices.

  • Described in any language
  • Up to 3 takes
  • Private

Transcription

Audio or video up to 5 GB and 8 hours in; timed segments and subtitles out.

  • SRT · VTT · TXT
  • Speaker labels
  • Auto language

Streaming

Audio starts playing in about two seconds while the rest is still being generated. Built for voice bots and call centres; Vietnamese for now.

  • First audio ≈ 2 s
  • PCM or WAV
  • 8–48 kHz

API & SDK

Zero-dependency SDKs for TypeScript, Python and Go. Every file is a signed URL.

  • TypeScript · Python · Go
  • Signed URLs

Safe by default

Clips and transcripts are private. One key per app, rotate without downtime. Every token spent shows in your usage.

  • Private clips
  • Key rotation

Voice library

165 voices in 10 languages, Vietnamese first.

  • AdamMale · youthful, husky, podcast, social-video
  • Duyên Hà MyFemale · low, slow, steady, calm, documentary, e-learning
  • GiangFemale · bright, clear, e-learning
  • Hoàng ĐặngMale · low, warm, documentary
  • HàoMale · clear, e-learning
  • Mai AnhFemale · northern, expressive, lively, news, e-learning

Use cases

Built for teams that publish audio.

Developers

From text to audio in one call.

Generate speech, episodes and transcripts from your own backend.

  • Speechclips up to 5,000 characters, any of 165 voices
  • Long-formepisodes up to 100,000 characters
  • Transcriptionfiles up to 8 hours
npm install @vovix/voixa
import { Voixa } from '@vovix/voixa'

const vx = new Voixa({ apiKey: process.env.VOIXA_API_KEY })

const clip = await vx.say({
  voiceId: 'en-alba',
  text: 'The lighthouse keeper switched the lamp on at dusk.',
})

console.log(clip.url) // WAV, signed URL

Pricing

Start free. Pay as you grow.

Token details →

Starter

$4
≈ 3.4 hours of audio
200,000 tokens$0.02 / 1,000 characters

Try a project, a few videos or a short audiobook.

Request this pack

Creator

$12
≈ 14 hours of audio
800,000 tokens$0.015 / 1,000 characters

For regular videos, podcasts and lessons.

Request this pack

Studio

$65
≈ 103 hours of audio
6,000,000 tokens$0.011 / 1,000 characters

For dubbing and high-volume production.

Request this pack

Every pack includes all voices, voice cloning and design, podcasts, transcription and the API. Failed jobs are refunded in full.

Every account also gets free tokens every day, no card needed. Hours are estimated for Vietnamese speech, where one token is one character.

Give your team a voice today.

Create an account in a minute. Free tokens refill every day, no card needed.