Message-Coverage Auditor for Pitch and Media Rehearsal
A tool that records a practice speech and shows which key points were mentioned and which were forgotten.
A rehearsal tool that scores a recorded practice pitch or interview answer on message coverage: which of your intended key points actually got said, which were dropped, and where you drifted onto tangents.
Concept
A rehearsal tool for anyone who must deliver a fixed set of messages under pressure — sales pitches, media interviews, conference talks. Step 1: the user enters their intended key messages (the "what to say"), or imports them from a talk track or deck. Step 2: they record a practice run. Step 3: the tool transcribes the run and scores it on content coverage, not fluency: which intended messages were actually delivered, which were dropped, which got tangled, and where the speaker drifted onto tangents — with a timeline marking each topic shift. The paper frames content selection as a distinct stage that precedes phrasing (the "what-to-say problem"), which is why fluent speakers can still fail a pitch: the phrasing is fine, the content selection is wrong. Topic-drift detection is grounded in discourse research showing that topics are systematically initiated, sustained, and dropped in naturally occurring conversation, so drift is a detectable, structured event rather than noise.
Why now
The paper explicitly separates content selection from syntactic form selection, giving a principled basis for scoring a spoken rehearsal on message coverage independently of delivery polish — something current speech-coaching tools do not do, since they count filler words and pace rather than whether the intended points landed. Discourse research on how topics are initiated, sustained, and dropped provides the analytic grounding for automatic tangent detection. With reliable ASR and LLM alignment between an intended message list and a transcript, the intended-vs-delivered comparison can now be automated per rehearsal run, making structured coverage feedback cheap enough for everyday practice rather than formal evaluation.
Written by GLM · 10 Oct 2026
AI assessment
A focused, high-utility tool for high-stakes communication that shifts the focus from delivery polish to content accuracy, leveraging LLMs to solve the 'what-to-say' problem.
- Evidence strength 3/5
- The research provides a theoretical framework for separating content selection from phrasing and defines topic drift, but the actual implementation relies more on modern LLM capabilities than the cited linguistic studies.
- Market pull 5/5
- There is a clear, urgent need in sales enablement and executive coaching for a way to ensure key talking points are hit without requiring a human coach for every rehearsal.
- Novelty & moat 4/5
- While 'Orphaned-Topic Detector' exists for support, applying this to proactive rehearsal and 'message coverage' for speakers is a distinct and valuable shift in user and intent.
- Feasibility 5/5
- With current ASR (Whisper) and LLMs (GPT-4), mapping a transcript against a list of key points and identifying tangents is a straightforward engineering task.
- Wedge clarity 5/5
- The 'Message-Coverage Auditor' is a sharp, specific tool with a clear entry point: auditing a recorded pitch against a talk track.
- Simplicity / focus 5/5
- The idea avoids platform creep, focusing exclusively on the comparison between intended messages and delivered speech.
Scored by Gemma 4 · 10 Oct 2026. Rated against a fixed rubric (evidence, market, novelty, feasibility, wedge, simplicity), relative to the closest ideas already in the catalog. A prior estimate to compare ideas before real-world signal arrives.
Persona discussion
AI personas trained on real people's expertise debate this idea as it evolves.
View the discussion →Act on this idea
Ideas only matter if someone runs with them. Your message goes straight to the founder's inbox — nothing is stored on our servers.
Who benefits
- Yoodlicompany
AI speech coach for pitches and interviews; message-coverage scoring differentiates it from competitors that only report filler words and pacing.
- Gongcompany
Revenue-intelligence platform for sales calls; a coverage module would show reps and managers which intended talk-track points actually got delivered on live calls.
- Speakeasy Inc.company
Executive communication coaching firm; objective coverage reports give media- and presentation-training clients measurable rehearsal feedback between coaching sessions.
- Toastmasters Internationalorganization
Members rehearse structured speeches against defined objectives; automated coverage and drift reports augment the manual speech evaluation process.
Research it builds on
- Topic as a Discourse Notion: a Study of Topic in the Conversations of Children and Adults( )Elinor Ochs Keenan, Bambi B. Schieffelin · 2026 · 287 citationsAll ideas from this paper →
- Conceptualizing and Formulating in Sentence ProductionGerard Kempen · 2026 · 117 citationsAll ideas from this paper →