Kiosk UI

On Blits.ai this is an agent with a knowledge base of approved airport content, retrieved with hybrid search, and a few custom functions that call the airport's flight information, directory and routing services as REST calls. A flow handles the fixed paths such as asking the traveler's language, and language detection and translation steps let one agent serve several languages.

The agent answers on the digital human channel, which streams a 3D avatar with lip synced speech to a browser on the kiosk screen, or on the web chat and WhatsApp channels, or through the REST and WebSocket API channels for a custom kiosk application. Voice uses streaming speech to text and text to speech from the providers on the platform.

Guardrails check input and output, PII masking runs at the gateway, and human handover passes the conversation to a live agent platform such as Salesforce, Freshdesk or Zoho SalesIQ. Test suites replay the top questions in each language on every change, monitors check the assistant on a schedule, and analytics show top intents, unanswered questions and channel use. The platform is model agnostic and offers EU and UAE data residency. Blits.ai does not supply the kiosk or hologram hardware.

On Blits.ai the associate copilot is an AI agent over a knowledge base of product, promotion and policy content with hybrid retrieval and document version control, with custom functions for live promotions, trade in values and the customer's account. Associates use it through the web widget on store tablets or in Microsoft Teams, and rich in chat cards show device and plan options that the associate can show the customer.

For customers, the same agent runs as a digital human: a photorealistic avatar with lip synced speech, streamed to an in store screen, with voice input and text as an alternative. Human handover escalates to a staff member, guardrails check answers against the operator's policies, test suites of real associate questions are rerun after every content change, and the platform is model agnostic. Run Blits.ai in our cloud in the EU, Qatar, Singapore, the UAE, the US or Australia, in your private cloud, or fully on premises.

On Blits.ai the guide is an AI agent grounded in a knowledge base built from the curated texts, documents and crawled pages of the organization, retrieved with hybrid search, plus a SQL knowledge base with the collection or points of interest records (identifier, title, location, opening hours). The organization's own app sends the visitor's stop, position or scanned object number with each message through the REST or WebSocket API channel, and custom functions call the collection system, ticketing or events calendar for live details.

The same agent runs in the embeddable web chat, WhatsApp and a digital human on a kiosk, and speaks through text to speech from many providers with custom voices, with TTS caching so fixed stop narration plays instantly. Multi language support detects the visitor's language and switches mid conversation. Guardrails keep answers inside the approved content, analytics show what visitors ask, test suites replay questions per stop before each change, and the platform is model agnostic. Run Blits.ai in our cloud in the EU, Qatar, Singapore, the UAE, the US or Australia, in your private cloud, or fully on premises.

On Blits.ai bots are multi language: each bot has a language list, flows hold localized content per language, the platform detects the resident's language and can switch mid conversation, and machine translation runs through Amazon, Google, IBM or Microsoft. An AI agent grounded in a knowledge base in the official language can answer in the resident's language, with strong Arabic support (normalization, Arabic voices and regional Arabic models).

On the phone, speech to text across nine providers and text to speech across thirteen let a voice agent converse in many languages, and self hosted transcription with language detection keeps audio on Blits.ai infrastructure. PII masking, guardrails and human handover to a staff member or interpreter complete the governed setup, and test suites check answers per language. Run Blits.ai in our cloud in the EU, Qatar, Singapore, the UAE, the US or Australia, in your private cloud, or fully on premises.

On Blits.ai the avatar is a digital human channel: a photorealistic 3D avatar streamed to the browser, with lip synced speech and an emotion engine that drives expression and voice. It is backed by an AI agent with a knowledge base of the institution's approved product, fee and policy content, retrieved with hybrid search, and by voice input with text as an alternative. Each branded persona runs as its own instance, and the same agent can also run in web chat and through the REST or WebSocket API in a mobile app.

Guardrails check input and output against the institution's policies, PII masking runs at the gateway, and human handover passes the conversation to a live agent platform, and a flow or custom function can book an appointment. Regulated steps follow a flow with deterministic checks, and any account action is a custom function behind the institution's own authentication. Test suites of real customer questions run after every content change, analytics show conversations and unanswered questions, and model agnostic routing lets the institution choose or switch the model. EU and UAE data residency is available for the hosting.