Web app
Inworld
Inworld AI provides realtime text-to-speech, speech-to-text, and LLM routing for building voice agents. It targets developers and enterprises needing low-latency, scalable voice AI.
- Platform
- Web app
- Categories
- Large Language Models (LLMs), AI Character, Game
- Indexed
- 2024-02-19
About Inworld
What is Inworld AI
Inworld AI is a realtime voice AI platform that offers text-to-speech (TTS), speech-to-text (STT), and LLM routing. It is designed for building realtime conversational agents with low latency (sub-130ms first-chunk for TTS Mini, under 250ms P90 for Max). The platform claims to be the #1 ranked realtime voice AI, based on user rankings on the Artificial Analysis Speech Arena. It supports voice cloning, voice design, and over 100 languages.
Key features
Inworld AI's key features include:
- Realtime TTS with sub-130ms latency and prices starting at $10 per million characters.
- Voice cloning from 15 seconds of audio, with cross-lingual support for 15 languages.
- Text-based voice design: describe accent, age, tone, and energy in natural language.
- Advanced voice direction: add bracketed instructions to adjust tone, speed, volume, and pauses.
- Realtime STT with pricing at $0.10 per hour.
- LLM routing that selects the best model and tools for each user and context.
- Speech-to-speech that understands, reasons, and interacts.
Use cases
Inworld AI is used across industries for:
- Voice-first companions: building ongoing, personal, emotionally engaging AI interactions.
- Agentic workforce: creating AI agents for customer support, scheduling, and more.
- Learning and education: language learning apps like Talkpal use Inworld for natural TTS.
- Health and wellness: virtual assistants for appointments and patient interaction.
- Interactive media: gaming and entertainment, with companies like Latitude and Isekai Zero using it for AI characters.
Pricing
Inworld AI has cut prices by half or more for most developers. Pricing details include:
- Realtime TTS: $10 per million characters (down from $100 for ElevenLabs).
- Realtime STT: $0.10 per hour (down from $0.46 for Deepgram).
- LLM routing: no markup (0% vs typical 5% for gateways).
- Dedicated GPUs: $5 per GPU-hour (down from $10+ for hyperscalers). These rates are for the Growth plan, with further reductions at enterprise scale.
Questions
Inworld FAQ
- What is Inworld?
- Inworld is an AI tool available as a web app in the Large Language Models (LLMs), AI Character, Game categories. Inworld AI provides realtime text-to-speech, speech-to-text, and LLM routing for building voice agents. It targets developers and enterprises needing low-latency, scalable voice AI.
- What is Inworld used for?
- Inworld is filed under Large Language Models (LLMs), AI Character, Game in the AITOOLIST index. Inworld AI provides realtime text-to-speech, speech-to-text, and LLM routing for building voice agents. It targets developers and enterprises needing low-latency, scalable voice AI.
- Is Inworld a web app, an extension or a native app?
- Inworld is a web app.
Listed in the AITOOLIST index since 2024-02-19.