Skip to main content

Upload a text transcript with speaker names

NEXT lets you upload written transcripts as recordings, just like audio and video files. This is useful when you already have a transcript of an interview, sales call, support conversation, or other customer interaction and want to analyze it in NEXT.

When a text transcript is uploaded, NEXT converts it into a transcript automatically. If the file includes speaker markers in the correct format, NEXT will also create speaker turns automatically so the transcript is ready for analysis without manually splitting speakers afterwards.

Supported file formats​

For text-based transcript uploads, NEXT supports:

  • Plain text files (.txt)
  • Microsoft Word files (.docx)

Format speaker turns​

To import speakers automatically, format each speaker turn like this:

[speaker:Interviewer] What do you think?
[speaker:Interviewee] I love it
[speaker:Interviewer] Thank you

To make this work reliably, follow these rules:

  1. Start each new speaker turn with [speaker:Name]
  2. Put the spoken text on the same line, after the closing ]
  3. Keep a space between ] and the first word
  4. Repeat the marker every time the speaker changes

Example:

[speaker:Interviewer] What do you think?
[speaker:Interviewee] I love it.
[speaker:Interviewer] Why?
[speaker:Interviewee] Because it is much faster than doing it manually.

Multi-line speaker turns​

You can also keep multiple lines under the same speaker. In that case, all following lines belong to the current speaker until NEXT finds the next [speaker:...] marker.

Example:

[speaker:Interviewer] What do you think about the workflow?
Does anything slow you down today?

[speaker:Interviewee] The process is clear overall.
The manual speaker splitting is the main friction point.

In this example, each block is still treated as a single speaker turn.

Best practices​

  • Use the exact marker format: [speaker:Name]
  • Use the full display name you want to see in NEXT, such as Interviewer, Customer, or Jane
  • Keep each speaker change explicit instead of relying on punctuation or layout
  • Prefer .txt for the cleanest import if you do not need Word formatting
  • Review transcripts exported from other systems before upload and convert each speaker turn into the NEXT marker format

What NEXT creates from your file​

During processing, NEXT converts the text into transcript words and speaker blocks. If speaker markers are included, those speakers are created directly from the file.

If a transcript is uploaded without speaker markers, NEXT imports the full transcript as a single speaker called Speaker 1.

Troubleshooting speaker imports​

If the upload works but speakers are not recognized, check the following:

  • The line starts with [speaker:
  • The speaker name is closed with ]
  • There is transcript text on the same line after the marker
  • Each speaker change has its own marker
  • The file is uploaded as .txt or .docx