Text Analysis, Essence or AI Chat: which to use
Text Analysis gives five fixed lines, the Essence gives a summary with timecodes, and the chat answers any question and works with your own text.
Updated November 2, 20262 min read
All three start from the same thing, the words in a recording. They differ in how much you get and how much you can steer the result.
Which one to pick
Text Analysis: a voice message or a short talk, when one screen is enough: topic, key points, tone.
Essence in Text Transcription or Video to Text: a lecture, a meeting, an interview, when you need a summary with chapters and timecodes.
AI Chat: when you want to ask questions, get a specific format (a reply email, a table, a list of decisions) or work with a text you already have.
How they compare
| Text Analysis | Essence | AI Chat | |
|---|---|---|---|
| Result | 5 fixed sections: Topic, Key Points, Tone, Speech Features, Summary | Brief, Key points, Chapters with timecodes; Q&A, terms and action items when the recording has them | any answer you ask for |
| Your own text | no field for it | no field for it | paste it into a message |
| Files | audio up to 100 MB, video, link up to 2 GB | audio up to 100 MB, video up to 3 GB in Video to Text, link up to 2 GB | up to 10 files, up to 50 MB each; a link goes into the message |
| Follow-up questions | no, but Discuss analysis moves it to the chat | no, but Discuss in chat moves the transcript to the chat | yes |
| Language of the result | the app language | the app language | whatever you ask for |
| Cost | recognition about 0.73 tokens per minute plus a fraction of a token | recognition about 0.73 tokens per minute plus a few tokens | depends on the model and the length of the conversation |
Analysis in the chat
When you attach an audio file in the chat, chips appear above the input field. Content analysis gives four sections: Topic, Key points, Tone, Summary. It has no Speech Features section, but you can keep asking about the recording in the same conversation. Video files only get the Translate video chip; for a video, ask in your own words. More in attaching files in chat.
How to combine them
A common path for a meeting: run it through Video to Text or Text Transcription, read the Essence, then tap Discuss in chat and ask for a follow-up email or a list of who does what. The chat gets the full transcript, so you do not pay for recognition twice.
Frequently asked questions
Which is cheaper for a 30-minute recording?
Recognition is the main cost in the tools: about 22 tokens for 30 minutes. Text Analysis and the Essence add very little on top. In the chat the price depends on the model you pick and how many messages you send.
Contact support
Describe the problem and we will reply by email. Add the file name or job ID if you have one.