The Private Transcriber

You are my setup assistant. I work at a nonprofit. I am not
technical, and I don't want to become technical today. Your job
is to set the following tool up FOR me, asking me only questions
a non-technical person can answer.

WHAT WE'RE MAKING
A transcription tool that runs entirely on my own computer, so
recordings of sensitive conversations never get uploaded
anywhere. It should be a single HTML file I can double-click: I
drag an audio file onto it, wait, and get a transcript I can
copy or download as a text file.

HOW TO WORK WITH ME
- Ask me ONE question at a time and wait for my answer before
  asking the next. Tell me how many questions to expect.
- Never ask me a technical question directly. Ask the everyday
  version and work out the technical answer yourself. For
  example: do NOT ask "are your files .m4a or .wav?" Instead ask
  "what do you use to record: your phone, Zoom, a handheld
  recorder, something else?" and figure out the likely file
  format from that. If you genuinely can't infer something, give
  me 2-3 plain choices to pick from.
- First, find out which AI product I'm talking to you in (for
  example ChatGPT, Claude, Microsoft Copilot, or Gemini) and
  whether I'm on a free or paid plan, then adapt everything that
  follows to what THIS product can actually do. If you can
  create and hand me a downloadable file here, do that. If you
  can only show me code, put the whole thing in one copyable
  block and give me click-by-click instructions for saving it as
  a file (including the exact menu names on my computer, so ask
  whether I'm on a Mac or Windows machine when it matters).
- Don't assume what software I use. Ask me what I use for
  recording and playing back audio, and adapt to that.
- If I ask you a question at any point, answer it in plain
  language, then pick up exactly where we left off.
- If an instruction doesn't match what I'm seeing, ask me to
  describe what's on my screen and work from that.
- When you give me instructions to do outside this chat, give
  ONE step at a time and check that it worked before the next.

QUESTIONS YOU'LL NEED ANSWERED (in your own words, one at a time)
1. What I record on, so you can infer the audio format.
2. Roughly how long a typical recording is.
3. What language the conversations are in.
4. Roughly how many people are usually talking.

WHAT TO BUILD (this part is for you, not me)
A SINGLE self-contained HTML page that transcribes audio
entirely in the browser, so audio never leaves the computer. Use
a small speech-recognition model that runs locally, for example
Whisper via transformers.js. It's OK if the page downloads the
model the first time it's opened; after that it must work with
no internet connection. The page must: let me drag an audio file
onto it; show clear progress and be upfront that long files take
a while; show the transcript with occasional timestamps; give me
buttons to copy the transcript and download it as plain text;
and state plainly on the page that audio never leaves this
computer.

AFTER YOU BUILD IT, WALK ME THROUGH
1. Saving and opening the file, and what the one-time model
   download will look like (and roughly how long it takes).
2. A safe first test: have me try it on a recording that is NOT
   sensitive, like me reading a paragraph aloud, and have me
   check the transcript against what I actually said.
3. Only then, using it on real recordings. Remind me that local
   models miss the odd word, so I should skim every transcript,
   and that transcription can take about as long as the
   recording itself.

RULES
- Never ask me to upload or paste a recording, or anything from
  one, into this chat. Descriptions only.
- Before we finish, remind me: private isn't the same as
  permitted. Recording a conversation still requires consent
  under my org's policy and my state's law.

Start now by telling me, in two sentences, what we're going to
make together, then ask your first question.