In short: Upload the documents an agent should answer from, and it answers from them, or admits it does not know rather than inventing something plausible.
Definitions
- Grounded answer
- An answer composed from a passage retrieved from your material, so it reflects the document rather than the model's general knowledge.
- Retrieval
- The per-turn lookup that finds the relevant passage. It starts while the caller is still speaking, so the passage is in hand when the reply is composed.
- Custom vocabulary
- Product names, place names and industry terms taught to the recogniser so they are heard correctly on a noisy line.
What it does
- Price lists, policy documents, product sheets, FAQs, scripts: uploaded, and used on the call.
- Answers are grounded in the retrieved passage, so they reflect the document rather than the model’s general knowledge.
- When nothing in the material covers the question, the agent says so and offers a next step instead of guessing.
- Product names, place names and industry terms can be taught to the recogniser, so they are heard correctly on a noisy line.
- Updating a document updates what the agent says. There is no prompt to rewrite.
How it works
Retrieval runs per turn and starts while the caller is still speaking, so the lookup is already in hand when the reply is composed. A miss is a hard block on answering rather than a soft hint, which is what stops a confident wrong answer.
Related questions
Does the agent answer from my documents?
A CallWhiz AI agent answers from the documents you upload: price lists, policy documents, product sheets, FAQs and scripts. Answers are grounded in the retrieved passage rather than the model's general knowledge, updating a document updates what the agent says, and when nothing in the material covers a question the agent says so and offers a next step instead of guessing.
What happens when it cannot answer?
It says so and hands to a person, with the context. It is built to admit a gap rather than invent an answer.