Question 1

What does an AI meeting notetaker actually cost to run yourself?

Accepted Answer

Far less than most people expect — usually a few cents per meeting. A notetaker is two API calls stitched together: a speech-to-text pass that turns the audio into a transcript, and a language-model pass that turns the transcript into a summary with action items. Speech-to-text is billed by audio minute, so a 45-minute meeting at a typical rate of around a cent per minute costs roughly a quarter to half a dollar. The summary is where people assume the money goes, but a full 45-minute transcript is only about eight thousand tokens, and summarising it on a cheap model costs a fraction of a cent. Across a normal month of meetings the whole thing lands in the range of a single SaaS seat, which is why building your own is attractive once you have more than a couple of users.

Question 2

Why is speech-to-text the biggest cost, not the AI summary?

Accepted Answer

Because transcription is priced by the minute of audio while the summary is priced by the token, and a meeting produces very few tokens relative to how long it runs. People speak at roughly 130 words a minute, so even a long meeting yields only a few thousand words — well under ten thousand tokens for the model to read. On a cheap summarisation model that input costs a fraction of a cent, and the summary it writes back is a few hundred tokens more. The speech-to-text pass, by contrast, has to process every second of audio and is billed accordingly, so it typically accounts for the overwhelming majority of the bill. This calculator splits the two out explicitly so you can see that the LLM summary is essentially a rounding error and the real lever is your transcription rate.

Question 3

Is it cheaper to build my own notetaker or pay for Otter or Fireflies?

Accepted Answer

It depends almost entirely on how many meetings you record. A SaaS notetaker charges a flat monthly seat fee whether you record two meetings or two hundred, while a do-it-yourself pipeline charges per meeting with no floor. That means the API route wins for light and moderate users and the SaaS route wins for heavy ones, and there is a break-even number of meetings where they cross. This calculator prints that break-even directly: below it, running your own transcribe-and-summarise pipeline is cheaper; above it, the flat subscription is the better deal. Remember the SaaS price also buys you integrations, speaker labels, search and a polished UI, so the honest comparison is cost versus the engineering time to match those features.

Question 4

How do I estimate the token cost of summarising a meeting?

Accepted Answer

Start from the length of the meeting and a speaking rate. Multiply the minutes by words per minute to get the word count, then multiply by roughly 1.33 to convert words to tokens — that is the input the summary model has to read. Add the length of the summary you want back, in the low hundreds of tokens for a tight brief with action items. Price the input at the model's input rate per million tokens and the output at its output rate, and you have the summary cost. For a typical meeting this is a fraction of a cent, so the token side is rarely worth optimising; the money is in the speech-to-text rate and the number of meetings, both of which this calculator lets you vary.

AI 会議メモ作成者のコスト計算ツール

お金の行き先 - 会議ごと

SaaS と比較して独自のシートを構築する - 損益分岐点

概要は無料です。あなたはマイクの代金を払っています

この計算機の仕組み

よくある質問

AI 会議メモ作成ツールを自分で運営するには実際にどれくらいの費用がかかりますか?

AI の要約ではなく、音声からテキストへの変換が最大のコストとなるのはなぜですか?

自分でnotetakerを作るのと、OtterやFirefliesにお金を払ったほうが安いでしょうか?

会議を要約するためのトークンコストを見積もるにはどうすればよいですか?