Who is behind this

About Open Voice Shield

Open Voice Shield is built and operated by TelecomsXChange Canada Inc., the company named as the seller on every invoice the platform issues. The same people write the code, run the platform, answer support and sign the contracts. There is no reseller in between and no third party operating it on our behalf.

The company

TelecomsXChange is a telecommunications company. Open Voice Shield is one of its products, with its own name, its own site and its own contract, sitting under the company the way a product brand sits under the business that makes it. When the footer says a TelecomsXChange product, that is the relationship it is describing.

Being a telecom company first shaped the product. It takes a SIP trunk rather than an app, authenticates by source address the way a carrier interconnect does, bills per second, and returns a verdict on the call rather than a dashboard about calls. Those are the habits of people who have run switches.

What the team does

We run one service. Calls arrive on a trunk you point at us, we record and analyze them, and we hand back a verdict with the evidence behind it, a check on whether the route answered honestly, a caller ID result and an estimated quality score for each leg. Everything else on this site exists to explain that or to get you connected to it.

One address answers all of it. There is no separate sales team or tiered support desk, and no ticket gets routed to a different company. Write to support@telecomsxchange.com and the people who built the product read it. The contact page says what to put in the first email.

The rules we publish under

A fraud product is easy to oversell, so this site follows five rules.

  • Prices are published. The per-minute rate, the per-call analysis fee and every plan fee are on the pricing page, and the platform serves the same figures over its own API. You do not have to ask what it costs.
  • Limits are published. What the product does not see, cannot decide and will not promise is written down, including in the white paper under what we have not measured. A page that lists only strengths is a brochure.
  • No accuracy figure. We publish no accuracy, precision or recall number, because we have not published one. The method, the corpus and the raw results go to anyone who asks. If you see such a figure attributed to us, it did not come from us.
  • Every MOS is an estimate. The quality score is derived from jitter and packet loss on the media, not gathered from listeners, so the site calls it estimated every time.
  • No number we have not measured. If a figure appears on this site, it came from the running platform or from a published standard, and we can say which.

What is open, and what is not

The service is commercial. Some of the things built around it are not: the desktop plugin described in the Omarchy post is open source, and the interfaces are documented and open to use. The REST API, the OpenAPI document, the agent interface and the evidence export are all documented and usable without asking us for permission. The export opens offline with no account, so your evidence does not depend on us still being here.

Security and procurement

If you are reviewing us as a vendor, start with the security summary and then the operational detail, which covers authentication, tenant isolation, where call content lives, retention and deletion, and what the audit log records. Both pages describe controls that exist today, not planned ones. For anything they do not answer, ask, and put the answer in the contract.

Questions

Frequently asked questions

Who operates Open Voice Shield?

TelecomsXChange Canada Inc., the company named as the seller on every invoice. The same team builds the product, runs the platform and answers support.

Is Open Voice Shield a separate company?

No. It is a product of TelecomsXChange Canada Inc. with its own name and its own identity, in the way a product brand sits under the company that makes it.

Why does the site publish no accuracy figure?

Because we have not published one. We share the measurement method, the corpus and the raw results on request, and until a figure is published there is nothing to quote.

Why is every MOS described as estimated?

The score is derived from jitter and packet loss on the media legs, not collected from listeners, so calling it a measured opinion score would be wrong.