---
title: "Text notes"
description: "Upload a TXT, DOCX or PDF file and run the same keyword, sentiment and entity analysis you get on a transcript, without paying to transcribe anything."
---

> Documentation Index
> Fetch the complete documentation index at: https://docs.speakai.co/llms.txt
> Use this file to discover all available pages before exploring further.

# Text notes

A text note is a document you upload instead of a recording: a transcript from another tool, survey responses, meeting notes, a report. Speak AI analyzes it the same way it analyzes a transcript, so you get keywords, sentiment, named entities, and AI Chat without paying to transcribe anything.

![The upload drop zone, cropped from the upload window. It invites you to upload or drop a file and lists every format it accepts, which includes the txt, docx and pdf documents a text note arrives as, alongside the audio and video types.](/help/media/uploads/uploads-text-notes.jpg)

## Upload a text file

Text files go through the same upload window as audio and video. Open **Upload**, drop the file in or browse for it, and Speak AI reads the text out of it and analyzes it.

These file types work:

- **TXT**, plain text
- **DOCX**, Word documents
- **PDF**, up to 50MB per file
- **PNG** and **JPEG**, read with OCR, for a scan or a photo of a page

Speak AI builds an insight panel beside the text. Select any extracted insight to highlight every mention of it. **Explore Insights**, beside the insights search bar, opens the [Explore page](/help/insights/explore/) for deeper analysis and visualization of the same note.

Break the text into paragraphs rather than one continuous block. Clear paragraph breaks help Speak AI detect topics accurately.

## Bring in transcripts you already have

Already transcribed a recording somewhere else, in Zoom or by a human transcriptionist? Analyze that text without paying for transcription twice:

- **Save it as a TXT or DOCX file and upload it.** You can then run [AI Chat](/help/ai-chat/), [sentiment](/help/insights/sentiment/), and named entity recognition on it exactly as you would on a media file.
- **Import in bulk with [CSV import](/help/uploads/csv-import/)**, one row per note. Each row becomes its own text note, which suits survey responses and other short-answer sets. This is the route to take when you have the text in a spreadsheet rather than in separate documents.

## If you only have the text on screen

There is no in-app editor to type or paste a note into. Paste your text into a plain text file first, save it as `.txt`, and upload that. For many short pieces of text at once, put them in a spreadsheet and use [CSV import](/help/uploads/csv-import/) instead of making a file for each one.

Questions about a specific file? Send us a message on live chat and we will take a look.

Want a hand setting this up? [Book a free consult](https://calendly.com/speak-ai/consult?utm_source=docs&utm_campaign=consult) and we'll do it together on your account.

---
Related: [Uploads](/help/uploads/) · [CSV import](/help/uploads/csv-import/)

Source: https://docs.speakai.co/help/uploads/text-notes/index.mdx