Control the entire interface by voice
We build sites and apps A–Z on Next.js only — with VoiceUI inside (navigation, forms, FAQ, guide). No WordPress, Webflow, or legacy. Start: Pilot from 12,900 PLN net (narrow site + voice).
VoiceUI Universal — demo on all devices
This site runs in Universal mode: iPhone, Android, Safari, and Firefox use cloud speech recognition. On desktop Chrome/Edge we use the faster Web Speech API. Microphone access is required.
Already on Next.js and shipping yourself? SaaS platform: account, plan, key — voiceui-platform.vercel.app
118+
defined commands
PL/EN
command dictionaries
2–3 wks
typical pilot
Greenfield
Next.js / React
VoiceUI in practice
Navigation, forms, FAQ, and accessibility — one coherent command system
VoiceUI in action
The recorded session shows the full interaction cycle: navigation between sections, form handling, command centre, and guided mode. On this demo site Universal mode is active — you can test by voice on iPhone and Android as well.
VoiceUI Universal — demo on all devices
This site runs in Universal mode: iPhone, Android, Safari, and Firefox use cloud speech recognition. On desktop Chrome/Edge we use the faster Web Speech API. Microphone access is required.
- Section navigation with a single voice command
- Dictation and editing of contact form fields
- FAQ and command centre panel control
- Guided mode for users who need TTS support
Commands: "play video", "stop video", "pause", "close video"
Operating principle VoiceUI
Five technical pillars that distinguish VoiceUI from chatbots and simple voice assistants — deterministic commands, immediate UI response, and full integration with product code.
Interface command layer
Every command maps to a specific UI action: section, form element, FAQ accordion, or help panel — without the ambiguity typical of conversational models.
Speech synthesis and confirmations
The system informs the user about the completed operation. Voice responses can be enabled or disabled without losing control functionality.
Parallel input handling
Voice complements mouse and keyboard — it does not replace them. The interface remains fully accessible for all interaction methods.
Bilingual intent recognition (PL/EN)
Synonyms, phonetic variants, and contextual matching in Polish (pl-PL) and English (en-US/en-GB), optimised for Web Speech API and cloud STT.
Next.js / React implementation
VoiceUI is embedded in the front-end architecture — not a plugin, not an iframe. Full control over performance, accessibility, and visual consistency.
Benefits business and UX
VoiceUI turns investment in accessibility and user experience into measurable outcomes: better UX, broader reach, and clear brand differentiation.
Impact visible from first contact
In the premium segment, the first seconds shape how the product is perceived. A voice layer signals mature UX and deliberate interface design.
Differentiation from competitors
Full voice control of the interface (not just chat) is uncommon in the Polish market — a strong sales and brand argument.
Higher experience quality
Shorter paths to content and actions translate into longer session time and better conversion outcomes.
Expanded audience reach
Accessibility for people with disabilities, seniors, and users who prefer voice interaction.
Technological brand positioning
The brand reads as intentional about UX and accessibility — not another template with a chat widget bolted on.
Content navigation efficiency
Voice eliminates repeated clicks and scrolling — especially on extensive informational sites.
Long-horizon investment
Voice interfaces are gaining ground. VoiceUI is a production-ready layer you can deploy on greenfield Next.js.
Support for users with limitations
VoiceUI delivers real digital independence — not a symbolic compliance checkbox. Guided mode walks through site structure, reads sections, and enables full operation without pointing devices.
Activation: "enable guide" — navigation with section descriptions and content readout.
Motor limitations
Full site operation by voice only — without mouse, touchpad, or keyboard.
Low vision and blind users
Navigation with TTS confirmations, logical section structure, and high interface contrast.
Senior users
Natural commands instead of complex visual navigation — shorter learning curve.
Independence in the digital environment
Users complete tasks without third-party assistance — a key accessibility standard in public and commercial services.
Accessibility as an architectural foundation
VoiceUI was built on the premise that the interface must be usable for every user. Accessibility is not a layer added after deployment — it is part of the design process from day one.
Operational independence
Users independently achieve goals on the site — without workarounds or keyboard shortcuts known only to experts.
Semantics and UI patterns
Aria-labels, focus states, contrast, and heading hierarchy — in line with inclusive design best practices.
Input complementarity
The voice layer extends — does not restrict — mouse, touch, and keyboard operation.
WCAG guideline alignment
Design and implementation account for accessibility requirements at every stage of the product lifecycle.
Implementation process
We deploy VoiceUI exclusively in new Next.js/React projects — with full control over architecture, code quality, and interface consistency.
Greenfield projects only
The voice layer is designed in parallel with UX and information structure. We do not integrate VoiceUI with existing CMS or legacy systems without front-end rebuild.
Discovery and requirements analysis
Mapping user groups, interaction paths, and command scope against business goals.
Voice intent design
Command dictionary, synonyms, and error scenarios — tailored to product structure.
UI and VoiceUI layer implementation
Production code in Next.js/React with performance and accessibility standards maintained.
Tuning, testing, and deployment
Speech recognition optimisation, command regression tests, and release to production.
Tech Stack
Commercial implementation: CodeProSite.com — Next.js / React deployments with embedded VoiceUI technology
Two ways to use VoiceUI
VoiceUI is not a chatbot — precise voice commands on your site. Choose self-serve SaaS or a full turnkey deployment.
SaaS: add the layer yourself — one script, dashboard, subscription. Deployment: full funnel fit — we build it for you.
SaaS platform
Add the voice layer yourself: account, project, public key and a site snippet. No waiting for custom development.
Starter · Pro 249 PLN/mo · 14-day trial
- Account, project and public key (pk_…)
- Snippet to paste on your site
- Management dashboard and plan limits
- 14-day trial — no long-term lock-in
- Ideal to start and test in production
Turnkey deployment
Full layer on greenfield Next.js/React: command dictionary, design, forms, guide mode and launch. We deploy — you get a finished product.
Pilot · full packages in pricing below
- Discovery, architecture and product dictionary
- Implementation + browser testing
- Standard or Universal (mobile) option
- Training and post-launch maintenance
- For teams that want full control without DIY
We deploy VoiceUI for you
Three engagement paths: a fast pilot, the voice layer alone on greenfield Next.js/React, or a full product with design and SLA. No WordPress retrofit — production quality from architecture.
Pilot
Validate VoiceUI on 1–2 flows before buying a full deployment.
2–3 wks · credited to full
- 1–2 voice scenarios
- Next.js / React greenfield
- Demo for leadership
- Report + full proposal
Layer only
VoiceUI embedded in a new app/site you build — or we build with you.
3–6 wks · Standard; Universal higher
- Command dictionary for your product
- Forms + navigation + FAQ
- Command centre + guide mode
- Handoff documentation
Full product
Site or app from scratch + VoiceUI, premium design, SEO, launch.
6–12 wks · Pro → Enterprise
- Greenfield UI + VoiceUI
- Pro / Enterprise packages
- Integrations, SLA, EN
- Training + maintenance
Who it's for
- E-commerce, finance, insurance, health — accessibility and EAA
- B2B SaaS and web panels — hands-free onboarding and forms
- Premium brands — differentiation and UX innovation
- Public / education — inclusive access to content
- Agencies — white-label delivery for end clients
Who it's not for
- Local micro-business “we want AI for 2–5k”
- WordPress / legacy without front-end rebuild
- Chatbot widget only, with no UI control
- Projects not ready for greenfield Next.js/React
Why organisations deploy VoiceUI
Accessibility and compliance
Voice + guide mode complements WCAG and supports users with impairments.
01
Market differentiation
A live demo of full voice UI control — concrete in a pitch and PR, not another AI slide.
02
Less friction on key flows
Hands-free navigation and forms — shorter paths to conversion in selected scenarios.
03
How client deployment works
A repeatable process from brief to production — on a new site or app, not a bolt-on to legacy chaos.
- 01
Brief and discovery
Section map, flows, commands, Standard vs Universal, privacy/STT risks.
- 02
Architecture + dictionary
Section IDs, PL/EN intents, command centre, error scenarios.
- 03
Implementation
Layer in Next.js/React, forms, FAQ, guide mode, browser tests.
- 04
Launch + maintenance
Deploy, training, SLA package, command growth after go-live.
VoiceUI Pilot
A paid, short pilot on a new greenfield site or app — before you invest in a full deployment. See VoiceUI on your product, not only on voiceui.pl.
Implementation pilot
1–2 key voice flows (e.g. navigation + form) on a Next.js/React product. A working result in 2–3 weeks — ready for leadership demos and user tests.
Investment
net · indicative range · typical scope
Up to 100% of the pilot fee can be credited toward Pro / Enterprise if full work starts within 30 days.
Pilot scope
- 1 discovery session + mapping 1–2 flows
- VoiceUI layer: navigation and/or form (up to ~40 commands)
- Standard mode (desktop) or Universal upgrade on quote
- Preview deploy + 1 revision round
- Post-pilot report + full deployment proposal
What you get
A working demo on your product (or greenfield landing), material for a business decision, and a full quote with part of the pilot credited.
Pricing VoiceUI
Greenfield Next.js only — full A–Z site with VoiceUI. Pilot from 12,900 PLN = narrow scope (landing / 1–3 sections + 1–2 voice flows). Full Pro from ~35k; Enterprise from ~80k after brief.
What’s included — in 3 lines
- SaaS from 99 PLN/moYou paste a snippet on your site. We do not build your website.
- Pilot 12,900 PLNNew, narrow Next.js site (1–2 voice flows) + VoiceUI — not a full 10+ section site.
- Starter / Pro / EnterpriseFull A–Z greenfield Next.js site + VoiceUI within package scope.
Indicative ranges · net · Next.js A–Z + VoiceUI
12,900 PLN is not a 10+ page corporate site. It is a Pilot: a new, narrow Next.js site with VoiceUI. Full scope after the brief. We do not take WordPress / Webflow / legacy.
- Number of sections and application screens
- Voice command scope and scenarios
- Forms, integrations, multilingual support
- WCAG requirements and guided mode
- Timeline and deployment criticality
iPhone, Android, Safari, Firefox · Web Speech + cloud STT hybrid
Starter Universal
A–Z site + VoiceUI
New A–Z site + VoiceUI on all devices (iPhone, Android, Safari).
A–Z site · range · net
5–6 weeks
- Everything in Starter Standard (A–Z site)
- Web Speech + cloud STT hybrid
- iPhone, Android, Safari, Firefox
- Speech recognition API configuration
- Privacy policy for cloud STT
- Cross-browser and mobile testing
Pro Universal
A–Z site + VoiceUI
Full multi-platform A–Z site — like the voiceui.pl demo in Universal mode.
A–Z site · range · net
7–9 weeks
- Everything in Pro Standard (A–Z site)
- Full mobile and Safari support
- Cloud STT + Web Speech fallback
- STT API usage monitoring
- 45 days post-launch support
- 1.5h training (desktop + mobile)
Enterprise Universal
A–Z site + VoiceUI
Enterprise: A–Z site/app, multi-device, SLA, STT limits — quote after discovery.
A–Z site · after brief · net
9–16 weeks
- Everything in Enterprise Standard
- STT SLA + API limits and alerts
- Multiple command languages
- Priority mobile support
- API maintenance included for 3 months
- Dedicated technical lead
Initial consultation without obligation. VAT invoice. Staged payment: 40% / 40% / 20%. Binding quote after brief acceptance.
Standard vs Universal — which to choose
| Aspect | Standard | Universal |
|---|---|---|
| Devices | Desktop (Chrome / Edge) | PC + iPhone + Android + Safari + Firefox |
| STT technology | Web Speech API (local) | Web Speech + cloud STT (hybrid) |
| Monthly API cost | 0 PLN | From ~50 PLN (traffic-dependent) |
| Response speed | Fastest | Very good (mobile: +0.3–0.8 s) |
| GDPR | Simpler (no audio on server) | Requires cloud STT clause |
| Best for | B2B desktop, budget, quick start | EAA, mobile-first, full accessibility |
Extensions and maintenance
Upgrade Standard → Universal
from 5 900 PLN
STT hybrid + mobile testing for an existing VoiceUI project
VoiceUI layer only (no A–Z site)
from 18 900 PLN
If you already have Next.js — we add voice, not a full site from scratch
Pilot (narrow site + voice)
from 12 900 PLN
A–Z narrow scope · 1–2 flows · 2–3 weeks · not full site · credited to full
Maintenance Standard
from 990 PLN/mo
Updates, commands, support without cloud API
Maintenance Universal
from 1 490 PLN/mo
Support + monitoring and speech recognition API cost
Pack of 10 commands
from 2 400 PLN
New voice intents post-deployment
WCAG audit + report
from 4 900 PLN
Audit scope agreed individually
VoiceUI vs chatbot — how they differ
| Aspect | Chatbot / assistant | VoiceUI |
|---|---|---|
| Interaction model | Conversation, text responses | Deterministic commands → UI actions |
| Interface control | Limited or none | Full navigation, forms, FAQ |
| Predictability | Depends on LLM model | Every command = specific action |
| Accessibility | Chat supplement | Guided mode + WCAG + voice |
| Command languages | Often translated | Native PL/EN intents (pl-PL + en) |
| Deployment | Widget / plugin | Embedded in Next.js architecture |
Martyn Nowakowski
Creator of VoiceUI.pl — the idea and build of a layer for full voice control of a web UI (PL/EN).
VoiceUI.pl is my product idea: full voice control of a web application interface — navigation, forms, FAQ, scrolling, and guide mode — as an equal channel next to mouse and keyboard, not a chatbot.
It grew from two needs: real digital accessibility for people with limitations, and differentiation for brands that want UX beyond a chat widget.
I see this as a direction for web interfaces: natural speech to drive the product UI, with accessibility as a design requirement, not an option.
Original UI command system
A complete layer: command → on-page action (navigation, forms, FAQ, guide) — in PL and EN.
Technology with social mission
Digital accessibility as a product goal, not a marketing add-on.
User inclusivity
Support for blind, low-vision, senior, and motor-impaired users.
Value for enterprises
UX and accessibility differentiation that supports brand image and conversion.
Native PL/EN command support
Commands, synonyms, and intents designed for Polish and English — not translated from Anglo-centric models.
Engineering standards
Every project built from scratch in Next.js/React — performance, WCAG, and maintainable code.
Support post-deployment
We ensure continuity of the voice layer — from initial consultation through command development, monitoring, and SLA packages scaled to product size.
Technical support
Direct contact with the implementation team and defined response time.
Scope-based pricing
Every project priced after requirements analysis — transparently, without hidden line items.
Maintenance and evolution
Updates, new commands, speech recognition tuning after production deployment.
Usage licence
The client receives rights to operate the deployment; core technology protected by a licensing model.
4–8 week timeline
From discovery to production — with pace adjustable to project criticality.
Initial consultation
First conversation without obligation — needs analysis and deployment scope recommendation.
Questions and answers
Collected information on technology, deployment model, privacy, and collaboration terms — for product, IT, and executive teams.
Book a consultation
Describe your product and whether you want Pilot, layer-only, or full deployment — we reply within 24 business hours with next steps.
Direct contact
codeprosite@gmail.com
You can operate the form by voice — e.g. "enter name John Smith", "enter email john@company.com", "send message".