nari-labs/dia
dia
A TTS model capable of generating ultra-realistic dialogue in one pass.
Builder
nari-labs
nari-labs • individual
Stars
19,243
Using upstream star count
Forks
1,680
Using upstream fork count
Open Issues
0
Activity Score
0/100
0 commits in 30d
Created
Apr 19, 2025
Project creation date
README Summary
Dia is a text-to-speech (TTS) model developed by Nari Labs that can generate ultra-realistic dialogue audio in a single forward pass. The model is designed to produce high-quality synthetic speech that closely mimics natural human conversation patterns and intonation.
AI Dev Skills
Unmapped
Text-to-Speech SynthesisNeural Audio GenerationDeep Learning Model DevelopmentSpeech Synthesis ArchitectureAudio Signal ProcessingDialogue GenerationVoice Cloning
Tags
Text-to-Speech SynthesisNeural Audio GenerationDeep Learning Model DevelopmentSpeech Synthesis ArchitectureAudio Signal ProcessingDialogue GenerationVoice CloningVoice DubbingAccessibility TechnologyEducationGaming Character VoicesSelf-hostedVoice AICloud APISingle-pass GenerationReal-time AIGamingAudioAudiobook ProductionInteractive Dialogue SystemsTextOn-premiseMedia ProductionConversational AI Voice GenerationGenerative AIEntertainmentAccessibility Voice SynthesisPython
Taxonomy
Deployment Context
Skill Areas
Recent Activity
Updated 4 months ago
7 Days
0
30 Days
0
90 Days
0
Quality
prototype- Quality
- medium
- Maturity
- prototype
Categories
Dev Tools & AutomationPrimaryInference & ServingOther AI / MLGenerative Media
PM Skills
Developer Platform
Languages
Python100.0%
Timeline
- Project created
- Apr 19, 2025
- Forked
- Mar 23, 2026
- Your last push
- 4 months ago
- Upstream last push
- 4 months ago
- Tracked since
- Nov 19, 2025
Similar Repos
pgvector cosine similarity · $0
Loading…