This page focuses on where Smallest AI (Pulse) fits in a production voice stack. For full setup steps, credentials, and API details, use the documentation link above.
Overview
Pulse is Smallest AI's speech-to-text model, built for real-time transcription at very low latency. With Bolna's Smallest AI integration, voice agents can transcribe caller audio on phone calls with minimal delay, keeping conversations feeling natural instead of laggy.
Pulse is ideal for teams that want a fast, cost-effective transcriber to pair with Smallest AI's synthesizer for a lightweight, low-latency voice stack end to end.
Features & Use Cases
Ultra-Low Latency
Optimized for real-time streaming transcription so agents can respond the moment a caller stops speaking.
Cost Effective
Lower pricing structure suited to high volume deployments where transcription costs add up fast.
Built for Voice Agents
Tuned for live phone audio rather than offline transcription, handling noisy lines and natural speech patterns.
Pairs with Smallest AI TTS
Combine with Smallest AI's synthesizer for a full lightweight speech stack optimized for speed end to end.
Use Case: High Volume Call Centers
Run large scale outbound or inbound calling operations where transcription latency directly affects conversation quality.
Use Case: Cost-Sensitive Deployments
Keep per-minute transcription costs low for high call volume use cases without sacrificing real-time responsiveness.