Baarcha · Dido

Tunisian TTS — text to speech & AI voice | Dido

Dido is Baarcha’s Tunisian text-to-speech (TTS) model: write in Derja, also called Tounsi, and turn your text into an AI voice. Choose an available voice and listen in the playground. Use it to prepare a voiceover, read a written message aloud or add spoken replies to your application through the speech API.

Try Dido ↗API documentation Pricing and limits

What you can try

Listen to a written Tunisian message before sharing it.
Prototype spoken customer-service replies that a person can review.
Generate voice drafts for a video or an interface, then check pronunciation and pacing.

Try it on your own example

Start with the example below and open the text-to-speech playground.
Choose a currently available voice and generate the audio.
Listen for the first and last words, proper names, numbers and French phrases.
For the API, read /api/v1/voices for the current voice IDs, then use /api/v1/audio/speech with model dido.

A starting example

الشارع مسكر اليوم على خاطر الأشغال، لازم نعملو دورة طويلة باش نوصلو للخدمة في الوقت.

An example to test, not a guaranteed response.

Archived Dido recording

What to check

Listen to the complete output: words can be missed or pronounced incorrectly. The recording below is an archived Dido sample, not a measurement of the current model. Available voices can change. Voice cloning requires permission to use the speaker’s recording and an account; check the live interface for availability.

Evaluation worksheet ↗

Tunisian text-to-speech API

Create an API key in the dashboard and set BAARCHA_API_KEY on your server. Replace VOICE_ID with a current voice from GET /api/v1/voices. This request saves the generated audio to speech.mp3.

curl https://baarcha.tn/api/v1/audio/speech \
  -H "Authorization: Bearer $BAARCHA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "dido",
    "input": "الشارع مسكر اليوم على خاطر الأشغال، لازم نعملو دورة طويلة باش نوصلو للخدمة في الوقت.",
    "voice": "VOICE_ID"
  }' \
  --output speech.mp3

API documentation ↗

Tunisian TTS and AI voice: common questions

What is Tunisian TTS?

Text to speech (TTS), or speech synthesis, turns written words into spoken audio. Dido is built for Tunisian Derja. Enter your own text in the playground, select a voice and generate a recording you can listen to.

Can I use Dido for a Tunisian voiceover?

You can generate an audio draft for a video, a spoken message or an app interface. Start with a short script and listen to the full result before editing it into your project. Check the pronunciation of names, numbers and any French phrases.

Does Dido accept Arabic script, Tunizi and French?

The speech API accepts Arabic-script Derja, Tunizi and French code-switching as input. These writing styles can produce different readings. Try your exact wording and compare the recordings, especially when a sentence changes language.

Is a Tunisian AI voice the same as a voice assistant?

Dido generates speech from text. A voice assistant also needs speech recognition, a model to produce a reply and an application that connects the steps. Fennec provides transcription; Hannibal can work with the resulting text. A TTS request alone does not run a phone call.