Dia

About Dia

Dia by Nari Labs is a 1.6B parameter dialogue-focused text-to-speech model. It excels at generating natural conversational speech with support for nonverbal sounds like laughter, sighs, and coughs. Dia supports multi-speaker dialogue generation and voice cloning from 5-10 seconds of reference audio, making it ideal for creating realistic conversations and character voices.

Key Features

Use Cases

How to Use Dia

1. 1

   Sign up free or open the demo

   Create a free TextToSpeechAI account to claim your starter credits, or open the no-signup demo to try Dia dialogue right away.

2. 2

   Select the Dia engine

   In the TTS dashboard choose Dia from the engine list. Dia is the dialogue-oriented, ultra-tier model with multi-speaker and voice-cloning support.

3. 3

   Write a dialogue script with tags

   Compose your conversation using [S1] and [S2] to mark each speaker turn, and drop in nonverbal tags such as [laughs], [sighs], [coughs], or (gasps) where you want natural reactions.

4. 4

   Generate the audio

   Click generate to send your Dia script to our hosted GPUs. Dia renders the two-speaker dialogue with turn-taking and your nonverbal tags into a single audio file.

5. 5

   Download or call the API

   Download the finished dialogue in your chosen format, or automate it by posting the same [S1]/[S2] script to the TextToSpeechAI API with your account token.

Dia API

Generate speech programmatically using the TextToSpeechAI REST API.

curl -X POST "https://api.texttospeechai.com/v1/generate/" \
  -H "Authorization: Bearer YOUR_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "text": "[S1] Hello there! How are you today? [laughs] [S2] I am doing great, thanks for asking!",
    "voice": "en_US-lessac-medium"
  }'

Frequently Asked Questions