47 voices
Curated, distinctive voices. Custom voice cloning available on Business+ plans.
31 languages
Native fluency, not translation. Code-switching across languages mid-sentence.
180ms latency
Real-time TTS for voice agents. Faster than human reaction time.
Emotion control
Tone, pace, emotion. Direct the delivery, not just the words.
Streaming API
WebSocket streaming for real-time applications. PCM, MP3, Opus outputs.
Speech-to-text
Speaker diarization, word timestamps, custom vocabularies. 95%+ accuracy.
Pricing
Token-efficient, with volume discounts and burn incentives.
47 voices, 31 languages
Emotion + style control
Diarization + timestamps
Sub-200ms voice agents
Pricing is illustrative. Final rates are governed by on-chain parameters and may vary based on network state.
Staking requirements
Tier-based access. Higher stakes unlock better economics and more capacity.
| Tier | Required stake | Access |
|---|---|---|
| Builder | 50M QUBIC | Voice TTS/STT enabled |
| Startup | 150M QUBIC | Realtime API, custom voices |
| Business | 500M QUBIC | Voice cloning, telephony |
| Enterprise | Custom | On-prem, custom voices |
Example
Drop-in compatible with the OpenAI SDK.
from aigarth import Aigarth
client = Aigarth(api_key="sk-aigarth-...")
audio = client.voice.synthesize(
text="Welcome to Aigarth. The future of compute grows on participation.",
voice="atlas",
language="en-US",
emotion="warm",
)
audio.save("welcome.mp3")Enterprise benefits
Everything in the standard tier, plus the things enterprises need.
- Custom voice cloning with consent verification
- HIPAA-aligned for healthcare voice agents
- PCI-compliant for payment use cases
- Telephony integration (Twilio, Vonage)
- WebRTC and WebSocket SDKs
- Brand-safety filters and content moderation
- On-prem deployment
- 24/7 on-call support