Powered by neural speech synthesis

Text-to-Speech
as a Service

Production-ready voice synthesis with streaming support, multiple voices, and multi-tenant isolation. Deploy in minutes, scale without limits.

Built for Production

Everything you need to add natural-sounding voice to your application.

Real-time Streaming

Start hearing audio before generation completes. WebSocket and SSE streaming deliver chunks as they are synthesized - no waiting for the full result.

Multi-Voice Library

Choose from a curated set of neural voices, each with distinct character and tone. Pre-loaded at startup for instant response times.

Multi-Tenant Isolation

Each organization gets a dedicated subdomain with isolated usage tracking. Tenant resolution happens at the edge - zero configuration needed.

API-First Design

REST, SSE, and WebSocket endpoints. Simple query parameters, standard HTTP responses, WAV output. Integrate in minutes with any language or framework.

Try It Live

Type something and hear it spoken. Requires an org subdomain for API access.

.vocyon.fastflowtech.ai

Simple, Powerful API

Three ways to synthesize speech - pick the one that fits your use case.

POST /v1/tts Synchronous - returns complete WAV
curl -X POST "https://acme.vocyon.fastflowtech.ai/v1/tts?text=Hello+world&voice=fantine" \
  -H "Authorization: Bearer $TOKEN" \
  --output speech.wav
GET /v1/tts/stream Server-Sent Events - chunked PCM audio
const source = new EventSource(
  `https://acme.vocyon.fastflowtech.ai/v1/tts/stream?text=${encodeURIComponent(text)}&voice=fantine`
);
source.onmessage = (e) => {
  const data = JSON.parse(e.data);
  // data.type: 'start' | 'audio' | 'end'
  if (data.type === 'audio') playChunk(atob(data.data));
};
WS /v1/ws/tts WebSocket - persistent connection, binary PCM chunks
const ws = new WebSocket("wss://acme.vocyon.fastflowtech.ai/v1/ws/tts");
ws.onopen = () => ws.send(JSON.stringify({
  text: "Hello world",
  voice: "fantine"
}));
ws.onmessage = (e) => {
  if (e.data instanceof Blob) playPCM(e.data);
  else console.log(JSON.parse(e.data));
};

All Endpoints

Method Path Description Auth
GET /health Health check and model status None
GET /voices List available voices None
POST /v1/tts Synchronous TTS - returns WAV JWT
GET /v1/tts/stream Streaming TTS via SSE JWT
WS /v1/ws/tts WebSocket streaming TTS JWT

Simple Pricing

Pay for what you use. No hidden fees, no minimum commitments.

Starter

For prototyping and small projects

Free
  • 1,000 characters/day
  • All voices included
  • REST API access
Most Popular

Pro

For production applications

$49 /month
  • 500,000 characters/month
  • Streaming + WebSocket
  • Priority processing

Enterprise

For high-volume and custom needs

Custom
  • Unlimited characters
  • Custom voice cloning
  • Dedicated infrastructure