Learn how speech-to-text and speaker diarization work, then dictate emails and prompts by voice, turn meetings and interviews into minutes, action items and themes, get help live on a call, and let a workflow file the action items for you.
Turn Talk Into Work: Voice and Meetings AI is a free, hands-on AI course for consultants, account managers, HR and operations people, and anyone who spends hours a week in calls and interviews or typing what they could say: 9 short modules (about 1.5 hours in all) where you do each step yourself in a safe practice copy of Wispr Flow, ChatGPT, Speak AI, CoHear and other tools, which checks your work. No coding, no cost, and you can start without signing up.
Who it’s for: Consultants, account managers, HR and operations people, and anyone who spends hours a week in calls and interviews or typing what they could say. No coding until the last two modules.
Talking is where most office work starts: client calls, interviews, team meetings and quick voice notes. Then it gets lost. Notes are patchy, the recap goes out a day late, action items sit in someone's notebook, and the insight from ten customer interviews never makes it into a decision.
This course starts with how voice AI actually works: speech-to-text turns sound into words, its mistakes land on names, numbers and jargon, and speaker diarization works out who said what. Then you use one tool per job, in order: Wispr Flow to write emails and prompts by voice, ChatGPT to turn a transcript into minutes and owners, Speak AI to pull themes and quotes from customer interviews, CoHear as an AI notetaker in the room, OLVA for live help on a call, and two n8n workflows that put every action item into a tracker.
The payoff: faster writing, same-day recaps, nothing dropped between meetings and research you can act on, with the checks that keep a transcription error from becoming a wrong decision. Every practice uses made-up recordings and meetings.
Who this course is for
The first five modules are beginner: a browser and the meetings you already have. The rest are intermediate: an AI notetaker, a live assistant and two n8n workflows.
You can already
Run or join meetings and calls
Use email, ChatGPT and a spreadsheet
Nothing else: no code and no accounts to start (the practice runs in your browser)
After this course you can
Explain speech-to-text, word error rate and speaker diarization, and what to check in any transcript
Write emails and AI prompts by voice and check them before sending
Turn a transcript into minutes, decisions and action items with owners and dates
Pull themes and quotes from customer interviews and check them against the audio
Use an AI notetaker and a live meeting assistant with everyone's consent
Send meeting action items to a tracker automatically with n8n
What you’ll build
Module 3: You help Maya Chen, who runs the office at Northgate Accounting, a made-up firm, answer a client email and write a ChatGPT prompt by voice.
Module 4: From the transcript of a team’s Monday meeting: a five-line summary, the decisions, an action items table (task, owner, due date) and the open questions.
Module 5: You upload a customer interview to Speak AI, read the transcript by speaker, scan the keywords, people and sentiment, and ask AI Chat for a summary with quotes and fixes.
Module 9: A recap with decisions, an action items table and open questions. Four rows in a Google Sheet called Kickoff Action Items with the columns Meeting, Owner, Task, Due date and Status. And a five-slide kickoff deck.
Welcome to Stepthrough: you learn AI by doing it, on practice copies of real app screens, free and with no risk to your real accounts. Your first practice takes under two minutes.
Client calls, interviews, team meetings and voice notes hold decisions that nobody writes down, or that get written up a day late from memory. Voice AI captures all of it as text, then turns that text into minutes, action items and themes.
Most office work is writing things you already know: replies to clients, quick updates, prompts for ChatGPT. Typing them out, fixing typos and adding punctuation eats a big part of the day.
After every meeting, someone spends half an hour writing up what was decided. Or nobody does, and the tasks quietly get lost until someone asks "wait, who was doing that?"
Small teams run customer interviews and sales calls, then nobody has time to listen to them again. The problems customers name never reach the people who could fix them.
Record a team training session on your iPhone with CoHear, check the AI notes and action items against the transcript, make a one-page handout and share it behind a password.
Give OLVA your client notes, get private answers and fact checks during a Zoom call, then turn the meeting into minutes, a follow-up email and a PowerPoint deck.
Build an n8n workflow that reads your meeting notes from a sheet, has AI pull out who agreed to do what by when, and adds one row per task to an Action Items sheet.
You just ran a kickoff meeting. You have a page of rough notes and three jobs: a recap with owners and dates, a tracker your team can tick off, and a short deck for the client.
For the practice: a web browser. Each module runs in a copy of the real tool’s screens, so you need no account and no sign-up to start module 1.
To do it for real: a ChatGPT account, a Speak AI account, an n8n account and a Gamma account. Each module says which one it uses.
Time to complete
About 1.5 hours of hands-on practice across 9 modules (244 steps in all). Modules take 2 to 15 minutes each, so you can do one in a sitting and come back for the next.
The ideas behind each module
Each module teaches one idea you will use again and again. Here it is in plain words: what it is, why a business uses it, and a real example.
Speech-to-text writes the most likely words with timestamps. Word error rate measures its mistakes. Speaker diarization splits the transcript by voice, and you name the speakers.
Why a business uses it
Every voice tool starts here, and its errors land on the names, numbers and dates that matter most. Knowing that tells you what to check.
A real example
An HR consultant's transcript turns "the Q4 renewal" into "the queue for renewal", and diarization labels the client as Speaker 1 until she names him.
AI for Meetings · The whole meeting, before, during and after: research briefs, live help and follow-ups.
Questions about this course
Is Turn Talk Into Work: Voice and Meetings AI free?
Yes. All 9 modules of Turn Talk Into Work: Voice and Meetings AI are free, and you can start module 1 without signing up.
How long does Turn Talk Into Work: Voice and Meetings AI take?
About 1.5 hours of hands-on practice across 9 modules, beginner to intermediate. Each module takes 2 to 15 minutes, so you can do one at a time.
Which tools will I practice in?
Wispr Flow, ChatGPT, Speak AI, CoHear, OLVA and n8n. You practice in a copy of each tool's real screens that checks every step, so nothing touches your real accounts, and no coding is needed.
Who is Turn Talk Into Work: Voice and Meetings AI for?
Consultants, account managers, HR and operations people, and anyone who spends hours a week in calls and interviews or typing what they could say. No coding until the last two modules.
Where do I start?
Start with module 1, Start Here: How Stepthrough Works. The modules are in order, so each one builds on the last.
How does AI transcription work?
A speech recognition model listens to the recording in short slices and writes the most likely words with timestamps, then adds punctuation. Speaker diarization splits the result by voice. Accuracy depends most on clean audio.
What is speaker diarization?
The step that works out who spoke when. It groups speech by how it sounds and labels it Speaker 1, Speaker 2. It does not know names, so you add them, and it slips when people talk over each other.
How accurate is AI transcription?
Accuracy is measured as word error rate, the share of words that are wrong. On clear audio it is usually high, but the errors cluster on names, numbers, dates and jargon, so check those before you send minutes or quote anyone.
Is it legal to record a meeting with AI?
Tell everyone the call is recorded and get their OK. Some US states require everyone on the call to agree, and client contracts may add rules. This course uses only made-up recordings.
Do I need to code?
No for the first six modules. The last two use n8n, which is visual but needs connected accounts and an API key.
What do I need before I start Turn Talk Into Work: Voice and Meetings AI?
Only a web browser for the practice: it runs in a copy of each tool's screens, so you need no account and no sign-up to start module 1. To repeat a module in your own tools afterwards, the modules list a ChatGPT account, a Speak AI account, an n8n account and a Gamma account.