Realtime TTS-2 is OpenAI's text to speech model. A real-time text-to-speech model from Inworld built for low-latency streaming voice generation in interactive applications.
Specifications
Canonical IDopenai-realtime-tts-2
TypeText to Speech
StatusActive
CreatorOpenAIOpenAI
Providers
Input ModalitiesText
Output ModalitiesAudio
Benchmarks
Elo Rating
#27

Capabilities

Input1/5
Text✓
Image·
Audio·
Video·
PDF·
Output1/5
Text·
Image·
Audio✓
Video·
Embedding·
Capabilities0/13
Reasoning·
Adaptive Reasoning·
Function Calling·
Parallel Function Calling·
Structured Outputs·
Native JSON Schema·
Web Search·
URL Context·
Computer Use·
Code Execution·
File Search·
Prompt Caching·
Assistant Prefill·

Pricing by Provider

US Dollar ($)
Per 1M tokens
ProviderStandard
Audio In
$ / 1K chars
Replicate logo
Replicate
replicate/inworld/realtime-tts-2
$0.025

Cost Calculator

US Dollar ($)
Preset:

Versions

VersionReleasedContextInput / 1MOutput / 1MStatus
Realtime TTS-2————Current
Inworld Realtime TTS 2————Available
Realtime TTS-2 Research Preview————Available
Soniox TTS 2 RT————Available
Step TTS 2————Available
StyleTTS 2————Available
TTS HD 2.5————Available
VoxCPM2————Available
TTS 1—$15.00—Deprecating
TTS 1 HD—$30.00—Deprecating
AssemblyAI TTS 1————Available

Model IDs

openai-realtime-tts-2
realtime-tts-2
replicate/inworld/realtime-tts-2