SLNG.
Advanced

Websocket Access for High-Volume STT & TTS

Overview

For customers with high-volume or real-time requirements, SLNG offers websocket-based access to our Speech-to-Text (STT) and Text-to-Speech (TTS) models. Websockets enable low-latency, persistent connections ideal for streaming audio, live transcription, and scalable voice AI applications.

We support custom websocket integrations for both STT and TTS. If you need to process large volumes of audio, require real-time streaming, or have specialized workflow needs, our team can help design a websocket solution tailored to your use case.

Note: Websocket endpoints are not enabled by default. Please contact us to discuss your requirements and enable access.

What You Can Request

  • ✅ Websocket Access to TTS Models (e.g. real-time speech synthesis, continuous audio streaming)
  • ✅ Websocket Access to STT Models (e.g. live transcription, streaming recognition)
  • 🌍 Regional Routing (e.g. run websocket sessions in specific regions for compliance or latency)
  • 🛠 Custom Configuration (e.g. max session length, concurrency, audio format, authentication)
  • 🔒 Dedicated Capacity (reserved resources for mission-critical workloads)
  • 📈 High-Volume Scaling (support for thousands of concurrent sessions)

How to Request Websocket Access

Websocket endpoints are provisioned on demand for qualified use cases. To get started:

1. Contact Us (Existing Customers)

  • Reach out to your SLNG account manager
  • Or message us in your shared Slack channel

2. Email (New Users)

Email:
📧 [email protected]
Subject: Websocket Access Request: [brief description]

Please include:

  • Your use case (brief)
  • STT, TTS, or both
  • Expected volume (sessions per day, concurrency)
  • Region preference (if any)
  • Latency or compliance requirements
  • Audio format or protocol needs

Example Request (Email)

Code
 
Subject: Websocket Access Request: Real-Time TTS for Call Center

Hi SLNG team,

We operate a call center platform and need real-time TTS for dynamic voice prompts, with support for 100+ concurrent sessions. Low latency is critical. Ideally, sessions should run in the EU for compliance.

Please advise on next steps for enabling websocket access.

Thanks!
— Team AcmeAI

What Happens Next

  1. 🧠 We'll review your requirements and assess technical fit.
  2. 📬 You'll receive a response within a few business days.
  3. ✅ If eligible, we'll provision websocket endpoints and provide integration details.
  4. 🔄 We may suggest alternatives or optimizations if needed.

Why Websockets?

Websockets are ideal for:

  • High-throughput, low-latency audio streaming
  • Real-time transcription and synthesis
  • Persistent, bidirectional communication
  • Scalable, event-driven voice AI

Quick Links

Last updated: June 2025

Last modified on
Request a model