A customer-service team already has a Foundry-hosted chat model that summarizes written transcripts. They now want to summarize recorded phone calls instead of typed chats, starting from the raw audio recordings. What must they add to their solution to make this possible?
Select an answer to reveal the explanation.
Short Explanation
The summarizer they already have is great at reading text, but a phone call recording is not text - it is sound, and no chat model can read sound directly no matter how capable it is. What is missing is a translation step: something that listens to the recording and turns it into a written transcript, which the existing summarizer can then chew through exactly like it does with typed chats today. Making the chat model itself bigger or giving it a longer memory window does not help, because the problem was never about how much text it can read, it is that there is no text yet. Turning the audio into number-vectors for similarity search solves a search problem, not a convert-sound-to-words problem. And tightening the safety filter does not touch the audio at all - it screens content after the fact, it does not create it. The missing piece is a dedicated transcription model sitting in front of the summarizer.
Full Explanation
The correct answer is B. The existing chat model summarizes text, but raw audio recordings are not text, so a speech-to-text model must first transcribe the recordings into written form; the chat model can then summarize that transcript exactly as it already does for typed chats. Option A is incorrect because a longer context window only changes how much text a model can read at once, it does not give a text-based chat model any ability to interpret audio input, so the recordings would still be unreadable to it. Option C is incorrect because an embedding model converts input into numeric vectors for similarity search or retrieval, not into readable transcript text, so it does not solve the need to produce a summary. Option D is incorrect because content safety filtering screens generated or submitted content for policy violations; it has no role in converting audio into text and would not enable summarization even if applied. Because typed chat text and recorded audio are fundamentally different modalities, the pipeline needs a dedicated transcription step before the existing summarization step can work on call recordings the same way it already works on chat transcripts.