Voice Agent API is unclaimed —

Voice Agent API

Build realtime voice agents on one API, with the benchmark voice model and a bill that fits on one line.·simbavoice.ai

Voice Agent API is a developer platform for building realtime, two-way voice conversations with an AI agent, bundling LLM reasoning, speech-to-text, text-to-speech, and telephony orchestration into a single per-minute rate. It is powered by the Simba 3.2 voice model, which produces first audio in under 100ms and streams natively. Agents can take inbound calls on a provisioned number or a customer's own SIP trunk, make outbound calls, or run over WebRTC in a browser, using Python and TypeScript SDKs or a REST endpoint.

What it's for

realtime voice agent conversationsinbound phone calls on a provisioned numberinbound calls on a customer's own SIP trunk (BYOC)outbound callsbrowser-based voice agents over WebRTCtool calling and knowledge base lookups during callsbatch calling campaigns

Features 20

1,500+ voices across 30+ languages
Batch calling campaigns (Scale plan)
Bring-your-own-model (BYOM) support (Scale plan)
Browser-based voice agents over WebRTC
Forward Deployed Engineer support (Enterprise)
Inbound calls on a customer-owned SIP trunk (BYOC)
Inbound calls on a provisioned phone number
Knowledge base and tool calling (Pro plan)
Outbound calling
Python and TypeScript SDKs
Realtime voice agents bundling LLM, speech-to-text, text-to-speech, and telephony orchestration in one per-minute rate
REST API endpoint (api.speechify.ai/v1/agents)
SOC 2 Type II certification
Speech marks (word-level timestamps) for captions/highlighting
SSML and emotion control with 13 emotion presets
SSO and signed DPA (Enterprise)
Streaming playback, up to 20,000 characters per request
Sub-100ms first audio from a streaming-native voice model (Simba 3.2)
Tool kinds: builtin, webhook, client, MCP
Voice cloning from a 10-30 second sample

At a glance

free tierAPI

TypeWeb app

DeploymentCloud

PlatformsWeb

Fordevelopers, solo devs and early-stage projects, teams shipping voice in production, enterprises with regulated or high-volume traffic

CompanySpeechifyAI

Integrations

LiveKitVapiDeepgramTwilioStripe

Complianceself-reported

SOC 2 Type II

Resources

Pricing

Free plan is permanent (not a trial) with hard caps (50K TTS characters/mo, 60 voice-agent min/mo) and no overage billing. Paid plans (Starter, Pro, Scale) bill overages from a prepaid balance at published per-character/per-minute rates with optional auto top-up. Enterprise is custom/quote-based with voice-agent rates starting from $0.06/min.

FreeFree

No credit card required; usage pauses at cap, no overage billing.

50K TTS characters/mo hard cap; 60 voice-agent min/mo hard cap; 3 concurrent calls; BYOC phone only

Starter$10/mo

Month-to-month, no commitment.

1M TTS characters + 120 voice-agent min included; 1 phone number; 6 concurrent calls

Pro$99/mo

Month-to-month, no commitment.

3M TTS characters + 1,200 voice-agent min included; 5 phone numbers; 12 concurrent calls

Scale$499/mo

Month-to-month, no commitment.

10M TTS characters + 6,000 voice-agent min included; 20 phone numbers; 30 concurrent calls

EnterpriseQuote

Custom/quote-based pricing; can be structured as annual agreements.

Custom phone allocation and concurrency; custom volume & rate commitments

Enterprise plans can be structured as annual agreements with volume discounts.