Dia Text to Speech
Dria directly generates realistic dialogue from transcripts. Audio conditioning enables emotion control. Produces natural nonverbals like laughter and throat clearing.
Input
Enter your dialogue with speaker tags and emotions
Result
Preview and download generated audio
Sample Audio
Powerful, Expressive Speech
Dia TTS combines advanced voice features with open-source freedom
Natural Dialogue Synthesis
By using speaker tags like [S1] and [S2], Dia TTS produces fluid, realistic dialogues that capture the nuances of human interaction.
Emotional Tone Control
Dia allows you to adjust the speech's tone and emotion through text cues or by providing reference audio samples, enabling expressive and contextually appropriate speech outputs.
Non-Verbal Cues
Enhance the realism of synthesized speech with non-verbal sounds. Dia supports cues like (laughs), (sighs), and (coughs), adding depth and authenticity to the generated audio.
Voice Cloning
Personalize the voice output by cloning specific voices. By conditioning on reference audio, Dia can mimic a speaker's unique characteristics, providing tailored and consistent voice generation.
Open-Source Accessibility
Dia TTS is freely available under the Apache 2.0 license. Access the model, code, and pre-trained weights on GitHub and Hugging Face, allowing for customization and integration into various applications.
Transform Your Projects with Natural Dialogue
Discover how Dia TTS can enhance your creative and technical projects
Voiceover for Film, Games & Animation
Bring your characters to life with voices that feel real. Dia supports multiple speakers, emotional nuance, and non-verbal cues like (laughs) and (sighs) — making it easy to create rich, believable dialogue. Whether it's a game, short film, or story concept, you get studio-quality voices without the studio.
Conversational AI & Voice Assistants
Dia brings conversations to life with natural flow, emotional tone, and multi-speaker support. Whether you're building a chatbot, voice companion, or interactive agent, it helps you create voices that feel more like a real conversation.
Podcasts & Audiobooks
Create lifelike narrators, craft dynamic character dialogues, and add non-verbal details to elevate your story. Whether you're an indie creator or looking to automate long-form content, Dia delivers top-quality results without compromise.
Frequently Asked Questions
Everything about Dia text to speech
Dia is a 1.6 billion parameter open-source text-to-speech (TTS) model developed by Nari Labs. It generates highly realistic, expressive speech from text, including natural dialogue, emotional tone, and non-verbal cues like laughter or sighs.