Wholesale Voice · Private AI Contact Centers

One kitchen.
Two outputs.

We buy voice capacity from regional operators in 100+ countries and sell it two ways: as raw carrier-grade minutes to enterprise buyers, or as turnkey private AI contact centers deployed on our own GPU infrastructure. Same plumbing. Two products. One operational team.

100+
Countries served
Regional
Operator partnerships
Tier-1
Termination carriers
A100
On-prem GPU inference

Wholesale voice and private AI — both, together

We are not an AI startup. We are not a pure wholesale broker. We are both — because in this market the two businesses reinforce each other.

The metaphor

A restaurant of the future.

To serve a meal people cannot find anywhere else, you need three suppliers no one else combines under the same roof — and the skill to bring them together on one plate.

🥩
The meat supplier
Wholesale voice capacity from regional operators — the foundation of everything we serve.
🥬
Fresh, GMO-free vegetables
Open-source AI engines — Ultravox, Whisper, Llama, Coqui — clean, inspectable, no vendor lock-in.
🌶️
Rare spices from India
Our unique touch: private deployment on our own GPUs. A flavor almost no one else on the market offers.

Put them together, and you get a dinner you will not find anywhere else — a wholesale voice business and a private AI contact center, built on the same kitchen, served by the same team.

Wine and dessert will come later — but that is a separate conversation.

1
Our operating core

Wholesale voice traffic

The business that pays the bills. We aggregate voice capacity and DID numbers from regional telecom operators and resell them to Tier-1 carriers, CPaaS platforms, and large enterprise buyers at competitive wholesale rates.

  • Buy from regional operators across MEA, SEA, LATAM
  • Sell to Tier-1 carriers, CPaaS, enterprise telecom
  • Supply Agreements + Service Agreements, USD wire, Delaware law
  • SIP redundancy via Telnyx (primary) and Twilio (failover)
2
Our premium product

Private AI contact centers

Built on top of the same carrier-grade voice infrastructure. For enterprise clients who want more than minutes — a managed AI voice service deployed privately on our GPUs, not in anyone's public cloud.

  • Ultravox v0.7 + Llama 3.3 70B on our NVIDIA A100
  • Whisper (STT) and Coqui/Piper (TTS), self-hosted
  • Seamless AI ⇄ human operator handoff
  • For regulated industries: healthcare, banking, government, legal

Wholesale voice — our operating core

The wholesale business is mature, high-volume, and grounded in regional carrier relationships we have spent years building. It funds the company and gives every downstream product carrier-grade voice lines.

Regional operators

Local minutes, DIDs, SMS termination from carriers across Africa, Middle East, South-East Asia, Latin America.

GoIPpro

Aggregation, SIP interconnect, quality monitoring, billing, settlement. Carrier-grade routes built from the regional supply.

Enterprise buyers

Tier-1 carriers, CPaaS platforms, large contact-center operators, fraud-prevention companies, DID aggregators.

What we buy

Voice minutes, DID number pools, and SMS termination capacity. Primary coverage across Africa, the Middle East, South-East Asia, and Latin America — markets where local presence and regulatory relationships matter most.

What we sell

Carrier-grade aggregated routes, E.164 number pools, and termination into Tier-1 networks. Buyers range from established carriers to CPaaS platforms to contact-center operators needing reliable volume at wholesale pricing.

How we contract

Standardized Supply Agreements with regional suppliers and Service Agreements with enterprise buyers. Settlement exclusively in USD via bank wire. All commercial contracts governed by Delaware law.

Private AI contact centers — what makes ours different

The AI voice market is crowded. ElevenLabs Agents, Vapi, Retell, Bland, Voximplant — all strong products. All running on someone else's public cloud. Great for marketing, lead-gen, non-regulated support. Not an option for regulated industries.

The rest of the market

Public-cloud AI voice

Every well-known AI voice brand runs on a hyperscaler. That works beautifully for outbound sales, lead qualification, and casual customer support. It breaks the moment a hospital, bank, law firm, or government agency tries to adopt it.

Customer voice and transcripts leave the perimeter. Shared GPU pools. Vendor controls the model and the terms. Compliance requires heavy contractual work. For regulated buyers the conversation usually ends at "our data cannot leave our data center."

GoIPpro

Private AI on hardware we own

Same class of engines — Ultravox v0.7 for speech-to-speech, Whisper for transcription, Coqui and Piper for synthesis, Llama 3.3 70B as the reasoning backend. But they run on our NVIDIA A100 in our own data center. Nothing in the call path leaves our rack.

We started with one A100 (already enough for production latency at first-tenant volumes) and will scale to three or four cards as demand grows. For clients who want premium cloud voice quality, ElevenLabs is available as an opt-in API integration — never as a mandatory part of the stack.

Products and pricing

Three products, transparent commercial terms. Wholesale is ready to trade today. AI is in MVP — shipping to first private-deployment clients this quarter.

Product 1 · Live now

Wholesale Voice Traffic

From $0.008 / min
Inbound and outbound voice minutes, DID numbers, SMS termination — resold from regional operators through our carrier-grade infrastructure.
  • Carrier-grade routes across 100+ countries
  • Tier-1 termination with Telnyx + Twilio redundancy
  • Monthly invoicing in USD via bank wire
  • Prepaid balance top-up via card (minimum $500)
  • Delaware-law contracts, standard Supply & Service Agreements
Start selling traffic →
Product 3 · Custom enterprise

Custom Engineering

Quote per scope
Bespoke delivery: multi-region deployments, regulated-industry hosting (HIPAA / PCI / government), integration with your CRM, outbound dialer programs, custom voice cloning.
  • Scoping workshop and fixed-price engagement
  • On-site or your-data-center deployment possible
  • Dedicated engineering team through go-live
  • SLA-backed managed operation
  • Milestone invoicing, USD wire or card up to limit
Request a quote →

All invoices issued by Goippro LLC (Delaware, USA). Card payments through Stripe for pilots and prepaid balances. Wire transfers in USD for production contracts.

Three deployment models — you choose, not us

Different regulations, different industries, different comfort with the cloud. We engineer all three paths with the same operational playbook.

Cloud-Grade

Fastest launch

Fully-managed voice AI on shared cloud infrastructure. Same class of engines the market already uses. Go live in days.

  • Sub-500 ms end-to-end latency
  • Pay-as-you-go per-minute billing
  • Best for marketing, lead-gen, non-regulated support

Hybrid / Convergent

Best of both

Sensitive turns (identity, payment, medical) stay on our private GPU. Routine turns (FAQ, directions, hours) run on cost-efficient cloud inference.

  • Policy-based per-turn routing
  • Up to 60% cost reduction vs pure private
  • PII never touches the cloud path

The full stack — carrier-grade voice, state-of-the-art AI

Our wholesale side runs on FreeSWITCH, SIP interconnects with Telnyx and Twilio, and direct partnerships with regional carriers. Our AI side runs on top of that same telephony, adding Ultravox, Whisper, and GPU inference. The full picture:

Speech → Speech core

Ultravox (self-hosted)

Sub-500 ms speech-to-speech. End-to-end voice model running on our A100. We pair Ultravox with Llama 3.3 70B locally, so the reasoning backend is also in our infrastructure — full data sovereignty.

Ultravox v0.7GLM 4.6Llama 3.3 70B
Speech → Text

Whisper (self-hosted)

OpenAI's Whisper large-v3 deployed locally. 70+ languages with automatic detection. Used for real-time transcription, call analytics, post-call summaries, RAG retrieval — all without transcripts leaving the rack.

Whisper large-v3Faster-Whisper
Text → Speech (self-hosted)

Open-source TTS on our GPUs

Coqui XTTS v2 and Piper run directly on our inference node. Natural barge-in: agents can be interrupted mid-sentence and recover gracefully. ElevenLabs is available as opt-in cloud API for clients who want premium voice.

Coqui XTTS v2PiperElevenLabs (optional API)
Telephony

FreeSWITCH + SIP partners

Carrier-grade SIP infrastructure. Redundant interconnects via Telnyx (primary Tier-1) and Twilio (failover). Plus direct trunks to our regional wholesale suppliers. PSTN numbers in 100+ countries, full E.164.

FreeSWITCHTelnyxTwilio
Orchestration

jambonz + control plane

Open-source CPaaS (jambonz) for the SIP/RTC layer. Our custom decision-gate handles handoff routing, session resume, multi-tenant isolation. Everything we run, we can modify — no black boxes.

jambonzCustom control plane
RAG & analytics

Private vector DB + LLM

Client knowledge bases ingested into Qdrant on our infrastructure. Retrieval-augmented generation runs entirely on-premise. Post-call analytics, sentiment, topic clustering — all private, all auditable.

QdrantLlamaPrivate RAG

The voice AI provider landscape — we know all of them

Every serious voice deployment in 2026 is a combination of carrier, AI engine, orchestration, and contact-center front end. Here is how the market actually lays out, and what we compose for clients.

Provider Role in our stack Latency Voice quality Cost profile When we use it
Ultravox
self-hosted
Speech-to-speech core on our A100★★★★★
<500 ms
★★★★GPU capex + opsOn-prem and hybrid deployments. Client data never leaves the rack.
Whisper
self-hosted
Speech-to-text, 70+ languages★★★★★GPU shareAlways, for analytics, transcripts, RAG, post-call summaries.
Coqui XTTS / Piper
self-hosted
Open-source text-to-speech★★★★★★★★GPU shareDefault TTS for private deployments.
ElevenLabs
cloud API only
Premium cloud TTS, conversational agents API★★★★
~1 s cascade
★★★★★$0.08–0.10/minOpt-in for clients who want premium voice and can accept cloud. Production elsewhere at Revolut, Klarna, Deliveroo, Deutsche Telekom.
Telnyx
primary carrier
PSTN/SIP Tier-1 — our main voice wholesale path★★★★Low wholesalePrimary telephony both for wholesale resale and AI call termination. Tier-1 global DIDs.
Twilio
failover carrier
Redundant PSTN/SIP, Flex UI★★★★20–30% above TelnyxAutomatic failover when Telnyx degrades. Flex for clients already on Twilio's platform.
Voximplant Kit
optional
Turnkey cloud contact-center★★★SaaS per-seatFor fully cloud-grade turnkey setups without data-sovereignty requirements. Native Ultravox integration.
jambonz
self-hosted
Open-source CPaaS, SIP/RTC★★★★★OSS + opsOur orchestration layer. Hosts the Ultravox app, manages transfers, bridges to human operators.
Vapi / n8n
as needed
Workflow orchestration and rapid prototyping★★★★LowMulti-agent orchestration, proof-of-concept builds, CRM and ticketing integrations.

Model-agnostic and carrier-redundant by design. The only non-negotiable: on-premise deployments run core inference on hardware we own.

Ultravox, Whisper, and what we learned running them in production

We are not an Ultravox reseller. We are engineers who have read every page of their documentation and every GitHub issue. We chose it for two reasons: sub-500ms speech-to-speech latency, and the ability to swap the underlying LLM for a model we fully control.

Why speech-to-speech, not the old cascade

Traditional voice AI cascades three separate models: STT, LLM, TTS. Every stage adds latency, and the conversation feels robotic. Ultravox v0.7 (GLM 4.6 base, late December 2025) compresses this into a single end-to-end model — the agent perceives tone, interrupts gracefully, responds in under half a second.

Model versionv0.7 / GLM 4.6
Turn latency<500 ms
Barge-inNative

Swapping the brain: Llama 3.3 70B, locally

Ultravox's strength is the speech layer, but the reasoning backend is swappable. We re-bind the audio encoder to Llama 3.3 70B on our own A100 — an open-weight, industry-grade model we own, with no per-token cloud billing and no third-party exposure of conversation content.

LLM backendLlama 3.3 70B (own)
Text-model flag--text-model
Audio-encoder flag--audio-model

Whisper for listening, Coqui and Piper for speaking

Whisper large-v3 runs locally on the same A100 — 70+ languages, automatic detection, real-time factor under 0.3×. Speech synthesis via Coqui XTTS v2 and Piper on our own hardware. ElevenLabs is available as an optional API integration for clients who want its premium voice, but it is never part of our self-hosted path.

STTWhisper large-v3
TTS primaryCoqui XTTS / Piper
TTS premiumElevenLabs (API)

Self-hosted ≠ fragile

Ultravox has real limitations every serious engineer reads about first: documentation gaps, memory accumulation under long sessions, inconsistent tool-calling, hardware-specific audio issues. We addressed each: session windowing, GPU memory watchdogs, tool-call schema validation, pinned kernel-level audio pipeline. The result runs unattended in production.

MemorySession recycling
Tool callsPre-validated
RAGQdrant, self-hosted

Start selling or buying traffic — today

Our wholesale business is already live. Tell us what you have and what you need, and we will come back within one business day with a routing plan and an offer.

This form takes 60 seconds. Every field helps us match you with the right route or buyer faster.

We reply within 1 business day from sales@goippro.com. No marketing spam.
Thanks — your inquiry is queued.
Your email client will open with the details pre-filled. Send it, and we will reply within one business day.

Seamless AI ⇄ Human handover

The hardest problem in voice AI is not the AI — it is the graceful handover when the AI reaches its limits. We built our handover architecture from the principles published in Ultravox's seamless-hybrid design and refined it in live production.

🤖

AI agent handles the call

80%+ of routine inquiries: order status, appointment booking, FAQ, lead qualification, payment reminders.

👤

Human takes over with full context

Operator receives conversation ID, full transcript, reason for handoff, concise summary. Customer never repeats themselves.

What triggers a handover to a human operator

Customer explicitly asks for a human
Sensitive intent detected: complaint, payment dispute, legal
AI confidence score drops below threshold
Same question asked and failed twice
Business rule: VIP customer, high-value transaction
Schedule rule: outside business hours route differently

Example: how a real handover looks

C
Caller
Hi, I'm calling about order 48291, it's been two weeks and nothing arrived.
AI
AI agent
Let me pull that up… I see order 48291 was shipped April 15th but flagged as lost in transit by the carrier. I can open a refund request right now, or arrange a replacement — which would you prefer?
C
Caller
This is the third time this has happened. I want to speak to a manager.
↪ Trigger: sensitive intent ("manager") + repeat issue detected → Decision gate routes to human with full context.
A
Human operator
Hi, this is Maria. I can see order 48291 was lost in transit and that you've had two previous issues. Let me personally escalate this — I'm offering a full refund plus a credit for the trouble.

Who we sell to — two audiences, one platform

Because the business has two products, it has two sets of customers. Both are enterprise-grade B2B. Both sign USD contracts under Delaware law.

Wholesale traffic buyers

Companies that want carrier-grade minutes and DID numbers
  • Tier-1 carriers — regional termination and DID fulfillment
  • CPaaS platforms — wholesale supply for Twilio/Bandwidth-style resellers
  • Large contact-center operators — multi-country voice capacity
  • MVNOs and aggregators — E.164 number pools
  • Fraud-prevention companies — high-volume call origination
  • Voice-broadcasting platforms — regional termination partners

Private AI contact-center buyers

Regulated industries where data cannot leave the perimeter
  • Healthcare providers — HIPAA-grade patient intake, scheduling, triage
  • Banks and insurance — KYC verification, claims, wealth-management front desk
  • Government agencies — citizen hotlines handling identity and benefits data
  • Law firms — attorney-client privileged intake and qualification
  • Regulated utilities — energy, water, waste — sectors with data-residency rules
  • Enterprise sales organizations — outbound pipelines where prospect data is competitive intelligence

The AI stack is proven — at planet scale

The engine class we deploy is already in production at the world's most demanding voice operations. Our differentiator: we install the same technology privately, without any of this data leaving your infrastructure.

Fintech
Revolut
4M+ customers
Voice support EU + UK, 30+ languages. TTR reduced 8×.
Fintech
Klarna
10× faster TTR
US first-contact AI agent for 35M customers.
Government
Czech national hotlines
~5 000 calls/day
Answer rate 60% → ~100%. 85% auto-resolution.
Healthcare
MyPlanAdvocate
~210 000/mo
Medicare pre-qualification. 2× conversion vs human-only.
Delivery
Deliveroo
75% success
Outbound voice to restaurants for order status.
Telecom
Deutsche Telekom
Strategic partner
National AI voice agent rollout announced Jan 2026.
Municipal
City of Midland, TX
3 000 calls/day
Bilingual EN/ES. −7 000 missed calls per month.
Solar
Freedom Forever
+90% efficiency
High-volume inbound for contractors and homeowners.
E-commerce
Kömpf24 "KIM"
−83% wait
Customer service at €5.48 per hour of equivalent dialog.

Questions we get asked most

Are you a telecom company or an AI company?
Both. Our core business is wholesale voice — buying capacity from regional operators and reselling to enterprise. On top of that infrastructure we offer a second product: private AI contact centers. The two businesses reinforce each other: our carrier relationships feed our AI product's voice paths, and our AI product is a premium service layered on our wholesale plumbing.
Does the AI business distract from the wholesale business?
No. They share infrastructure but have different sales motions. Wholesale is high-volume, commoditized, long contracts. AI is bespoke enterprise delivery with much higher margins per client. The wholesale side funds and feeds the growth of the AI side.
What if a client needs both wholesale minutes and an AI agent on top?
That is our sweet spot. One contract, one invoice, one operational team. The AI agent terminates onto the same SIP trunks we already manage for the wholesale traffic. It is the reason we built the company with both businesses from day one.
How do you compare to ElevenLabs Agents, Vapi, Retell?
They run in public cloud — excellent for marketing, outbound, non-regulated support. We run the same class of technology on our own hardware — designed for buyers who cannot use public cloud AI for regulatory reasons. When a client wants the cloud brands, we integrate them too (ElevenLabs is available inside our cloud-grade and convergent modes).
What hardware do you actually run?
An NVIDIA A100 (80GB) in our own data center. We started with one A100 — already enough to serve our first enterprise tenants at production latency — and as demand grows we will scale to three or four cards in the same cluster. Each enterprise tenant gets reserved capacity, not a share of a public pool. SIP connectivity is redundant via Telnyx (primary Tier-1) and Twilio (failover).
Who are you? What's the background?
Goippro LLC, a Delaware-registered company. Our team has years of background in wholesale voice — we came to the AI voice market as telecom insiders who saw a specific gap: every serious AI voice product runs in public cloud, which is fine for most buyers but unacceptable for regulated industries. We built the private-deployment version of what the market already loves.
How does billing work?
Wholesale: standard telecom settlement, monthly invoices in USD via bank wire, under Delaware law. AI: three models. (1) Per-seat — fixed monthly for a defined number of concurrent AI agents plus included minutes. (2) Per-deployment — flat fee covering dedicated GPU capacity and wholesale-priced minutes. (3) Custom enterprise for very high volumes, priced annually. All invoices from Goippro LLC (Delaware).
Can we audit your infrastructure?
Yes. For enterprise AI deployments we provide architecture documentation, security posture reports, and on-request data center tours. For highly regulated industries we can host in a facility of your choice with hardware we deliver and operate.

One platform. Wholesale voice, or private AI.

Whether you need a million voice minutes a month or a private AI agent for a regulated industry — talk to us. One operational team, one contract, one compliance posture.

sales@goippro.com →