Compare
Not the same VoiceShield: FraudCentral's monitor and TCXC Open Voice Shield, compared
TCXC Open Voice Shield and FraudCentral's VoiceShield share a word and a place in the call path, and do different jobs. VoiceShield, as FraudCentral publishes it, is a monitor: it listens to live calls and alerts a person. Open Voice Shield is a recorder and an enforcer: it records and transcribes every call, returns a verdict with the evidence attached, and acts on the flagged number's next call.
Names first
Two products, one word
Open Voice Shield is made by TelecomsXChange (TCXC), the wholesale voice and messaging marketplace, and lives at ovs.telecomsxchange.com. Its short form, OVS, is also the usual abbreviation of Open vSwitch, a virtual network switch that has nothing to do with phone calls. Write "TCXC Open Voice Shield" when the context could be read either way.
VoiceShield is one of four products of FraudCentral LLC, a Delaware company founded by people who ran intermediate carriers, alongside VoiceGuard, RoboDNO and RoboTools. FraudCentral says its products are used by over 300 carriers. There is also a Veridas "Voice Shield", a deepfake and liveness check for voice biometrics, which is a third, unrelated product.
The product you are comparing against
What FraudCentral's VoiceShield is, as published
Everything in this section comes from FraudCentral's own pages.
According to fraudcentral.io/voice-shield.php (read 30 September 2026), VoiceShield "sits between your switch and carrier, acting as an intelligent proxy" and "analyzes live voice interactions without interrupting or rerouting calls". It looks for known scam and traceback campaigns, robocalls and telemarketing calls, TCPA compliance issues and suspicious traffic patterns. The page describes it as informational monitoring: it raises alerts by e-mail and Discord webhook, and does not block traffic. It accepts up to 20 calls per second, works with any softswitch, costs $495 a month with 2,500 transcriptions included, and offers a 7-day free trial. The company states SOC 2 and ISO 27001 certification and publishes no accuracy figure.
Blocking is a separate FraudCentral product, VoiceGuard: set up as a carrier in the switch, it "blocks traffic based on media IP addresses", applies RoboDNO do-not-originate lists and stops double calls and ringless voicemail, deciding "in mere milliseconds", against a database the page puts at over 80,000 media IPs. It is priced by calls per second, from $500 a month for 250 to $1,000 for 1,000.
The two products suit a carrier, often an intermediate one, that wants to see scam and traceback campaigns on its network and block known bad media sources at high call rates, on a flat monthly price. Whether you need an alternative depends on what you need beyond that.
Questions that decide it
- Do you need the recording, the transcript and the red flags attached to each verdict, as evidence?
- Do you need the flagged number's next call blocked or diverted without a person in the loop?
- Are you a carrier with thousands of calls a second, or a business, platform or operator with specific trunks to protect?
- Do you want a flat monthly price, or to pay per call with no minimum?
- Do you need caller-ID verification, call quality scoring or a voice-agent gateway on the same hop?
Side by side
FraudCentral VoiceShield and TCXC Open Voice Shield, compared
VoiceShield entries are what FraudCentral's pages state; "not published" means the pages do not say. Open Voice Shield entries are the product as it runs today.
| FraudCentral VoiceShield | TCXC Open Voice Shield | |
|---|---|---|
| Where it sits | a proxy between the switch and the carrier, "without interrupting or rerouting calls" | one SIP hop between the switch and the carrier; relays the signalling and the media |
| What it listens to | live voice interactions | the whole call, recorded on separate caller and callee channels |
| When the verdict lands | during the call (real-time analysis) | minutes after hangup in standard mode; during the call in realtime mode, on request |
| What comes back | an alert by e-mail or Discord webhook | probability, category, impersonated entity, numbered red flags quoting the call, a TCPA note, the recording, the transcript, a PDF report and an evidence bundle |
| Action on the call | none: informational monitoring (blocking is VoiceGuard, by media IP and RoboDNO) | the flagged number's next call blocked or diverted automatically; a live call cut or diverted; alerts by e-mail, Slack and signed webhook |
| Caller-ID verification | not published | yes: a check proves what caller ID a route delivers, once or on a schedule |
| Call quality and FAS | not published | estimated MOS, jitter, packet loss per leg; false answer supervision detection |
| Voice AI agents | not published | route a call to your own OpenAI or xAI voice agent over the same hop |
| Capacity | up to 20 calls per second | licensed concurrent calls (ports) per account |
| Pricing | $495 a month, 2,500 transcriptions included; 7-day trial | prepaid: per analyzed minute plus a per-call analysis fee, no monthly minimum, $5 of credit at signup (pricing) |
| Data location | not published | Canada by default; United States on request, per account |
| Certifications | SOC 2 and ISO 27001 stated | none claimed; practices on the security page |
| Accuracy figure | not published | not published; the white paper says what was measured |
| API and agents | not published for VoiceShield (RoboTools has an API) | REST API for everything, an MCP server for AI assistants, a Claude plugin |
| Who it is built for | carriers, especially intermediate carriers facing tracebacks | carriers and operators, contact centres, platforms, and small and medium businesses with their own trunks |
How to decide
Which one fits your situation
Pick Open Voice Shield when
- a verdict has to come with proof: the recording, the transcript and the red flags, ready for a customer, an insurer or a regulator;
- the next call from a flagged number has to be stopped without anyone reading an alert first;
- you protect specific trunks (a business line, a contact centre, one wholesale route) rather than a whole carrier network;
- you want to pay per call, with no monthly minimum, and try it on $5 of credit;
- you also want caller-ID checks, quality scores or a voice-agent gateway on the same hop, or an AI assistant operating the account.
FraudCentral may fit better when
- you are an intermediate carrier whose first problem is tracebacks, and a monitor that alerts a person is what your process needs;
- you want media-IP blocking across all your traffic at hundreds of calls a second, which is VoiceGuard's job and not ours;
- a flat monthly price with a fixed transcription allowance suits your budget better than metered use;
- SOC 2 or ISO 27001 on the vendor's side is a hard requirement today.
What we do not claim
- No accuracy, precision or recall figure: none has been computed against labelled real calls.
- No self-serve real-time scoring: standard mode scores after the hang-up and acts on the next call; realtime mode is set up on request.
- No media-IP reputation database: Open Voice Shield judges the conversation, not the media source.
- No legal finding: the verdict is an analyst's read with evidence attached, and the TCPA note is not legal advice.
Both at once
They do not conflict
Both are SIP hops, so a switch can route through either or both. A common shape for a carrier: keep media-IP blocking in front of everything, and send the trunks whose calls you must be able to prove through Open Voice Shield, one call in ten to start, then read the verdicts. A business or platform with a handful of trunks usually needs only the second half. How a call travels through the hop.
Next step
Put one call through it
Create an account, point one trunk at the hop, place a test call with the scam sample and read the verdict with its evidence. Prepaid, per analyzed minute plus a per-call fee, $5 of credit at signup, no card.
Questions
Frequently asked questions
Is Open Voice Shield the same as FraudCentral's VoiceShield?
No. They are two products from two companies that happen to share a word. Open Voice Shield is made by TelecomsXChange (TCXC) and lives at ovs.telecomsxchange.com. FraudCentral's VoiceShield is one product of FraudCentral LLC, alongside VoiceGuard, RoboDNO and RoboTools. Neither is related to Open vSwitch, which is also shortened to OVS.
What is FraudCentral's VoiceShield?
As published on fraudcentral.io (read 30 September 2026), VoiceShield sits between a switch and its carrier as an intelligent proxy and analyzes live voice interactions without interrupting or rerouting calls, looking for known scam and traceback campaigns, robocalls, telemarketing calls and TCPA issues. It is described as informational monitoring only: it alerts by e-mail and Discord webhook and does not block. It takes up to 20 calls per second, costs $495 a month with 2,500 transcriptions included, and has a 7-day trial. FraudCentral's blocking product is VoiceGuard, which blocks by media IP address and RoboDNO lists.
What is the main difference between the two?
What comes back with a verdict, and what happens next. Open Voice Shield records every call on separate channels, transcribes it and returns a verdict with numbered red flags, a TCPA note, the recording and the transcript, and an automation rule can block or divert the flagged number's next call, or cut a live one. VoiceShield, as published, returns alerts for a person to act on and does not block.
Which one is cheaper?
It depends on volume. VoiceShield is a flat $495 a month with 2,500 transcriptions included. Open Voice Shield is prepaid per analyzed minute plus a per-call analysis fee, with no monthly minimum and $5 of credit at signup, so a low or variable volume costs only what it uses. The pricing page gives the per-minute rate and a worked three-minute call. Both publish their prices, which makes the arithmetic easy for your own volume.
Does either one block calls in real time?
FraudCentral's real-time blocking product is VoiceGuard, which blocks on media IP addresses and RoboDNO lists in milliseconds. VoiceShield itself does not block. Open Voice Shield blocks or diverts a flagged number's next call automatically, can cut or divert a live call, and in realtime mode, set up on request at a higher cost, scores the call while it is still up.
Can I run both?
Yes. They are both SIP hops, so a switch can route through either or both. A carrier might keep media-IP blocking for all traffic and put the trunks it must be able to prove through Open Voice Shield, for the recording, transcript and evidence bundle on each verdict.