Search Products

Stay Connected:
The cart is empty.
Cart contains {{shoppingcart.totalQuantity}} Item(s) Total: {{shoppingcart.subtotal}} {{shoppingcart.total}}

Browse by Category

Useful Links

Cloud vs Local Speechmatics for Court Reporters


Speechmatics can run in two very different ways: in the cloud or locally. Each option has strengths, limitations, and hardware considerations. This guide explains the differences so you can choose the setup that best supports your realtime work.


Cloud Speechmatics

Cloud Speechmatics sends audio to Speechmatics servers over the internet and returns text in realtime. This is the most common setup for reporters today.

Advantages:

  • No local installation required
  • Works on almost any modern laptop
  • Easy to update and maintain
  • Portable across locations and devices

Limitations:

  • Depends on courtroom Wi‑Fi or mobile hotspots
  • Latency varies with network conditions
  • Upload jitter can cause bursty or delayed output
  • Not ideal for restricted or low‑signal environments

Local Speechmatics

Local Speechmatics runs the engine directly on your hardware. This can be done through a dedicated appliance (often called a Boost Box) or, in some cases, installed directly on your laptop when licensed for local use.

Advantages:

  • No dependency on courtroom Wi‑Fi
  • Lower and more consistent latency
  • Stable performance in restricted environments
  • Ideal for high‑stakes realtime work

Limitations:

  • Requires strong CPU performance
  • Needs reliable cooling and fast storage
  • May require a Boost Box or local license
  • More technical setup than cloud

Which Option Is Right for You?

Cloud Speechmatics is convenient and flexible, but it depends heavily on network stability. Local Speechmatics offers consistent performance and lower latency, but it requires hardware capable of sustained processing.

If you are considering running Speechmatics locally—or pairing it with a Boost Box—M‑Tech can help you choose the right system for your workflow.

Learn about hardware requirements for local Speechmatics »