Back to openai-tts
OpenAI /

AI
TTS-1

API OnlyAudio & SpeechtextaudioUpdated November 6, 2023

TTS-1: OpenAI's Text-to-Speech Model

Model Overview

TTS-1 is OpenAI's text-to-speech API model, released in November 2023. It converts text input into realistic, natural-sounding speech audio in 6 distinct voices. Optimized for real-time generation with low latency, TTS-1 supports up to 4096 characters per request and outputs audio in multiple formats (MP3, Opus, AAC, FLAC, WAV, PCM). A higher-quality variant, TTS-1-HD, offers improved audio fidelity at higher latency for non-real-time applications.


✨ Key Features

Specification Table
FeatureDescription
6 VoicesChoose from Alloy, Echo, Fable, Onyx, Nova, and Shimmer voices
Low LatencyOptimized for real-time streaming audio generation
Multiple FormatsSupports MP3, Opus, AAC, FLAC, WAV, and PCM output
MultilingualGenerates speech in multiple languages following the text input
Two TiersTTS-1 (fast) and TTS-1-HD (higher quality) for different use cases

🔗 Resources

Specification Table

📜 License & Access

Proprietary — Available via OpenAI API.

You might also want to compare

Verified Sources

Model Specs

api-only

Parameters

Unknown

Context Window

Unknown

License

Proprietary

Deployment

api-only

Cost Tiers

TTS-1 HD

Resources & Links

Lineage

Model Family

Part of the openai-tts family

Only release in this line currently tracked.

Curator Notes

Bulk imported from OpenAI developer docs.

Compare Specs

Compare parameters, context windows, modalities, and benchmark scores of this model side-by-side with others.

Compare Model