Inbound voice agent with answers grounded in a knowledge base
Inbound calls concentrate on a small number of recurring requests, but they arrive in peaks and outside office hours. An operator answering repetitive questions is an operator not working on the cases that need judgment, and a call nobody answers is a request that is lost.
Analysis
Analysis of historical calls to establish which requests genuinely recur and which need a person, identification of the information required to answer and where it lives, and explicit definition of the perimeter: what the agent may say, what it must not commit to, and under which conditions it has to hand the call to an operator.
Solution
A voice agent that handles the inbound conversation and builds every answer on a knowledge base queried in real time, rather than generating it from the model's memory alone. Outside the defined perimeter, or when confidence is low, the call passes to an operator with the context already collected.
Architecture
Speech recognition and synthesis, RAG over the knowledge base to ground answers in verifiable content, guardrails on what the agent may state or commit to, explicit escalation rules and logging of every conversation turn for later review. The dominant constraint is latency: on a phone call a two-second pause is a broken conversation, and the architecture is designed around that budget before it is designed around the quality of any single answer.
Implementation
Started on a narrow subset of recurring requests, with real conversations reviewed and the perimeter widened only after checking behavior on edge cases and out-of-scope requests.
Duration
From kickoff to first production release: about one month.
Measured results
- 90% of in-scope calls completed without human intervention, with the remainder routed to an operator
- On the remaining 10%, the system correctly routed to an operator in 70% of the low-confidence cases, instead of answering without being sure
- Autonomous handling of recurring requests within the defined perimeter, with hand-off to an operator on out-of-scope cases
- Answers built on the knowledge base rather than on the model's memory alone, so they are updated by changing content instead of changing the system
- Continuity of response during peaks and outside office hours, where the alternative is an unanswered call
- Every conversation stays recorded and reviewable, including those that ended in escalation
ROI: Value is measured in recurring calls handled without operator time, fewer requests lost during peaks, and operator time concentrated on the conversations that need judgment.
Scope and measurement method
- Scope: a defined subset of recurring requests, with out-of-scope requests routed to an operator by design and not as a fallback.
- Evaluation set: real conversations reviewed and scored, including the calls where the agent correctly declined to answer.
- What is measured: correctness of the answers given, share of calls completed without human intervention, and escalation rate. The last is not a defect to minimize: a well-judged escalation is worth more than a risky answer.
- In voice, abstention matters more than in text: the listener cannot check the source while talking, and a wrong answer delivered fluently offers nothing to catch it on.
- Measured share: 90% of calls completed without human intervention, based on the real conversations reviewed and scored. Escalation rate: 70%, calculated on the low-confidence cases only (the remaining 10% of calls), not on the total. Observation period still to be confirmed.
Client: Project delivered for Aivora. aivora.ai/en/