Learning Cadence editorial, , updated , 6 min readHow we work
Learning Cadence combines measured speech feedback with a course and practice built around your goal. On suitable spoken exercises, you can see your pace, filler words, pauses and voice variation alongside feedback on what you said. The next attempt has a specific purpose: make the point clearer, support it better or give an important thought room to land.
A job interview, a wedding toast and a work presentation ask different things of a speaker. Cadence generates practice criteria from the situation and the teaching you have completed, so the content feedback can reflect what matters to that audience. This guide explains the checks and gives you a short rehearsal to try.
The short answer
Delivery measurements plus feedback on your message.
On suitable spoken practices, Cadence checks filler words, pace, pauses, vocal variation, sentence endings and length when a target is set. It also assesses your answer against criteria generated for the situation and completed lessons. You get feedback on your message and a separate delivery change to try. The measurements are estimates; content determines pass or retry.
Your next rehearsal
A speaking practice you can actually use.
Choose a real upcoming moment or a fictional one. Keep the situation the same when you try again, so you can compare the change you intended to make.
01
Describe the audience and outcome
Name who will hear you and what you want them to understand or decide. For example: I want to explain my contribution in a team-lead interview without taking credit for other people's work.
02
Learn the relevant skill
Build a course around that goal and complete the teaching you need. Course Practice uses completed lessons, and you can bring your own situation. Suitable spoken exercises let you rehearse aloud.
03
Read both parts of the feedback
Check the criteria for your message, then How it sounded. Look at the marked transcript to see where detected fillers and long pauses occurred, rather than relying on a count alone.
04
Change one thing and try again
Try a clearer opening or a deliberate pause before the result. Review whether that change helped the message you wanted to communicate. A willing listener can also tell you what they understood.
What the speech checks show you.
A recording can reveal a different problem from its written transcript. You may have the right words but rush the decision, leave a long gap inside a sentence, or let the last few words fade. Cadence makes the available measurements visible beside its coaching.
Filler words: detected words and phrases such as um, uh, like and you know, with a count and rate. Context rules help distinguish some ordinary uses from fillers.
Pace: words per minute and changes through the answer, so you can inspect how quickly you delivered it.
Pauses: longer gaps and where they fall in the answer. A pause before a key point can have a different purpose from a pause that breaks it apart.
Voice variation and endings: how much your voice rises and falls, and whether phrase endings fade, when the recording supports those measures.
Length: your speaking time against the exercise's target, when one is supplied.
The criteria change with the situation.
A rubric is simply the set of criteria used to review your answer. Cadence creates those criteria for the practice and the course material you have completed. Required criteria come from completed teaching; a later idea should not become a hidden requirement.
These are illustrative criteria, not fixed rubrics or recorded learner results. In an interview, useful checks can include answering the question, identifying your own contribution and supporting the result. In a wedding toast, they can include a considerate story, a connection to the couple and a clear wish. In a work presentation, they can include a recommendation, evidence separated from uncertainty and a decision request.
That gives the feedback something more specific to address than whether an answer sounded polished. A fluent interview answer can still omit what you personally did. A well-paced presentation can still leave the decision unclear.
Here is an original fictional example. You are presenting a pilot that reduced manual handoffs from five to three. Time savings have not been measured. You want two more weeks before a wider rollout. A useful content check asks whether your answer states the observed result, keeps the unknown visible and makes the decision request clear.
Suppose you communicate all three points but rush through the request. An illustrative delivery suggestion is: pause after the pilot result, then say the two-week request at a steadier pace. The content feedback and delivery feedback now point to different, practical changes. These example suggestions were written for this article; your feedback is generated from your attempt.
For your second attempt, keep the facts the same and work on that pause. Review the comparison with your previous measured answer, and consider whether the request is easier to follow. A lower filler count alone does not establish that the audience understood you better.
How Cadence turns a recording into feedback.
We make Cadence. Its speaking workflow turns your recording into words, measures delivery from the audio, and checks the response against the practice criteria before writing coaching. Pace, silence and pitch measurements come from the recording and transcript; the content evaluation checks what you said in the supplied situation.
Delivery coaching can use the exercise's speaking focus, such as leaving room for an important thought. The content result and delivery feedback stay separate. Speaking quickly or using a filler does not by itself make a correct answer fail, and typed words are not evidence of someone's pace, accent or confidence.
Lesson audio supports learning by listening. Recorded answers support speaking analysis. Together with personal course setup, completed-lesson grounding, saved practice attempts and Coach, these give you a connected route from learning an idea to rehearsing its use.
A delivery recording is capped at three minutes; rehearse longer talks in sections. Speaking feedback is available on suitable spoken practices, not every lesson or typed response. Live conversations have a separate daily allowance.
Transcription can miss words, and filler and hedge detection currently use English phrase lists. Word and pause positions are estimates. Available measures depend on the audio, so a zero filler count is not proof that you used none.
Cadence does not use a camera to assess eye contact or body language, diagnose speaking anxiety, or assign an overall eloquence or confidence score. The feedback can be wrong. Use it to choose a useful rehearsal, then check how the message works with a real listener when you can.
Put it into practice
Start it today, one lesson at a time.
If you want to keep going, Cadence turns what brought you here into a course with practice and saved progress. Learn at your own pace. Cadence is our app.
Yes. On suitable recorded speaking practices, Cadence detects fillers using the transcript and timing rules, shows a count and rate, and marks detected fillers in the answer. Detection currently uses English phrase lists and can miss words.
Does Cadence check more than filler words?
Yes. Available delivery measures include pace, pauses, vocal variation, sentence endings and length against a target. Separate content feedback checks your answer against criteria generated for the situation and completed teaching.
How is speaking feedback personalized?
Your goal and setup answers shape the course. Completed lessons and your practice situation shape the response criteria. The feedback refers to your answer, while delivery coaching can use the exercise's speaking focus and measured audio.
Can I use it for interviews, wedding speeches and work presentations?
Yes. Build a course around the situation and bring your own scenario to course Practice. On suitable spoken exercises, rehearse an answer or a section of up to three minutes. The criteria vary with the situation and completed teaching.
Should my goal be zero filler words?
Choose a communication goal you can check, such as making a request clear or explaining your own contribution accurately. Use detected fillers as one source of feedback and practice a useful change. A zero count by itself does not show that the message worked.
See it in concrete terms
Explore example courses for your goal.
Start with the speaking situation you care about, then make the course your own during setup.
