Ecosystem

Mind map

Mind map SVGMind map PNGMind map HTMLTrack map HTML

The person stays in the center. Voice, the recording, and the customer record sit beside the call. Emergency calling stays off the model arrow.

Integration map

Integration map SVGIntegration map PNGIntegration map HTML

One arrow is the live reply. Another runs from a recording to a transcript and a rubric. A third carries the handoff packet to a person. Drawing an arrow does not configure a vendor.

Match a sister to its job

Contact-center rooms

1. Words for this ecosystem
2. A mind map
3. Sister technologies
4. An integration map
5. Bridges across platforms
6. Future hooks

Glossary · One policy, three channels

Programming foundations

Sisters: Git, HTTP, JSON, unit tests.

A Python script can later call a local runner or a cloud API with the same JSON shape.

The same notebook becomes the capstone log.

How models work

Sisters: Tokenizers, Transformers, Decoding, Model cards.

The same next-token idea sits under chat, voice, and tool calls.

Pin a model id when you leave the toy explanations.

Prompting and evals

Sisters: JSON schema, Rubrics, Regression tests.

The same job-shape-test card works for chat, voice, and a tool result.

The capstone reuses this card.

Retrieval

Sisters: Search indexes, Embeddings, Rerankers, Citations.

A voice agent and a chat agent can share the same note index.

The capstone answers only from notes you are allowed to use.

Agents

Sisters: Workflows, State machines, Function calling.

The same loop can sit behind chat or a voice turn. The channel does not remove the cap.

MCP is one way to plug tools into that loop. It is the next track.

Model Context Protocol

Sisters: Plugins, Function calling, Language-server ideas.

A notes server can serve chat and a voice host if both speak the same protocol.

Treat the spec revision as something you pin, like a model id.

Voice and omni-channel agents

Sisters: Web Speech API, Turn-taking, Queues, CRM notes.

The retrieval index and the tool cap are shared. The channel changes the wait and the consent copy.

A later host could add a channel without rewriting the notes server.

Compute, cloud, and a capstone

Sisters: Schedulers, Regions, Key management, Risk frameworks.

A cloud endpoint can replace a local runner behind the same notebook and the same eval.

Pin the endpoint, the region, and the model id the day you check them.

Voice AI agent stack

Sisters: Speech-to-text, Language model, Text-to-speech, WebRTC, SIP.

A cascaded pipeline and a speech-to-speech model are two shapes the LiveKit docs name. The xAI Speech to Speech API is a hosted speech-to-speech session.

Pin the model id, the voice, and the date you read the docs.

Speech analytics and quality

Sisters: Automatic speech recognition, Sentiment analysis, Quality rubrics, Personal data.

The live voice pipeline answers the person. Analytics reads a recording later. Keep those clocks apart.

A later review can add a rubric you wrote. It cannot borrow a customer's tape.

AI agent safety

Sisters: Human handoff, NIST AI RMF, OWASP LLM risks, Recording statutes.

The voice channel does not remove the tool cap or the person who must approve an action that is hard to undo.

A later test can add a script you wrote. It cannot add a legal conclusion.

Contact-center ecosystem

Sisters: Workforce management, Call recording and quality, E911, Analytics, CRM, CPaaS.

The live reply, the recording, the customer record, the roster, and emergency calling are different systems. A bridge names the arrow. It does not merge the boxes.

Pin the page and the date. A hook in the notebook is not a promise this site will build the integration.