How are responsibilities separated in OMO Core?
The telephony layer receives and places calls. Audio processing manages the speech stream, while the conversational workflow controls questions and transitions. A separate business service validates actions. These boundaries keep media transport, model instructions and business rules distinct.
| Layer | Responsibility | Outside its responsibility |
|---|---|---|
| Telephony layer | SIP Channel, Distribution, Room, Participant and Media | A booking or CRM business rule |
| Audio processing | STT, turn completion, LLM, TTS and audio frame transport | Final source of data |
| Conversation workflow | Node, context, tool and transition | An unverified direct change to the database |
| Business service | Verification, availability, repeatability, secure performance and data retention | Transport of conversational audio |
| Operations panel | Configuration, sessions, auditing and control | Saving SIP or provider secret data in the browser |
Why does a node-based process increase accuracy?
Putting every task into one large instruction increases uncertainty. A node-based flow gives the language model only the context, expected fields and tools required at that stage, while each state transition is validated separately.
In OMO Core, a conversation is represented as a graph of connected stages. This approach is particularly useful for bookings, sales, surveys and identification, where conversation is natural but the boundary of action must remain clear.
Node is not just a textual question. It has entry condition, expected data, check, tool contract, success and recovery transitions. Therefore, the editing process in the panel and the actual behavior of the sound environment must be based on the same versioned model.
How does the system scale across multiple companies?
Multi-company voice AI scales when the number, SIP channel, process version, secret data, records and billing are separated by each workspace key. A voice handler can be launched from a common system image, but each call context is company-specific.
- SIP data is stored encrypted and never transmitted to the browser.
- A distribution rule associates a call with a specific workspace, number, and product.
- Processor concurrency is measured against processor, network, and vendor limits.
- Record, transcript and tool output do not cross company data boundary.
- Release version and process version per session are fixed for diagnostics.
Which protection rules should be built in?
Critical security rules must also exist outside of the agent's textual instructions: structured input, server authentication, data types, replay-safe execution, frequency limit, public address checking, audit event, and human-safe switching.
The NIST AI RMF considers risk management as part of the entire lifecycle, while the OWASP GenAI guidance focuses on the risks of instruction substitution and tool misuse. In voice AI, this means that what the user says is data, not system instruction.
OMO's approach to public safety does not claim to completely eliminate risk. The goal is risk visibility, a limited action surface, and that an uncertain response cannot create an unproven business action.
Official sources and further reading
Technical definitions are verified with primary sources. Links are opened in the official documentation of the corresponding project.
- Introduction to SIP — Telephony component documentation
A formal explanation of how SIP connects traditional telephony with real-time applications.
- Conversation graphs — Conversation component documentation
An official overview of the conversation graph, state management, and transitions between nodes.
- Top 10 LLM Risks of OWASP GenAI — 2026 — OWASP's Generative AI Security Project
2026 Current Public Guidance on Key Security Risks for LLM Applications.