The BotSurf voice API
Speech-to-text, text-to-speech, browser automation, IoT integration and AI agent voice control — one API, seven developer categories, each with its own dedicated page below.
AI & Machine Learning Developers
Add real voice interaction to models and agents, not a bolted-on wake-word demo.
5 developer types →Web & Browser Automation Developers
Control a browser by voice \u2014 navigation, testing and scraping without a keyboard in the loop.
5 developer types →IoT & Embedded Systems Developers
Voice recognition tuned for the memory and power budget real embedded hardware actually has.
5 developer types →Voice & Audio Application Developers
Speech-to-text and text-to-speech as a real API, not a demo that falls over past a sentence.
5 developer types →Enterprise & Business Application Developers
Voice for the tools a business already runs on \u2014 CRM, EHR, trading, and internal systems.
5 developer types →Platform & Infrastructure Developers
Voice as a layer other platforms build on top of, not just a feature inside one app.
4 developer types →API Consumers & Integrators
Add voice into a stack that already has several other APIs, tools and agents wired together.
3 developer types →Error handling
Automatic retries (2 attempts) with exponential backoff; configurable max_retries; default 60-second timeout, configurable per request.
Regional deployment
Data residency (your data stays in-region), private networking, and regional endpoints separate from global ones.
Framework compatibility
Free to start, usage-based pricing when you ship, or hold BOT in a connected wallet for API access with no separate subscription — the same holding already used for full app access.