Add the recording
Open ScreenApp, upload your audio or video and wait for processing. You can also start with an audio recording. Play a sample to confirm the source contains audible speech.
Check supported upload formats if the file is rejected. A supported extension still needs readable media inside the file.
Set the spoken language
Use the language spoken in the recording. Automatic detection is useful when it works; select the known language if detection is wrong. AI output language is a separate setting for generated writing.
The documented provider union covers 100+ languages. Coverage varies by processing path. The maintained list and provider evidence are on Accuracy and Methodology, rather than a separate list in this tutorial.
Review and correct
- Open the transcript and listen alongside it.
- Check names, numbers, specialist terms and unclear passages.
- Correct words using the transcript editing control.
- Check speaker labels against the voices you can identify.
- Save changes and export a fresh copy.
See transcript editing and speaker-label help. Do not turn unclear speech into a confident quotation.
Choose an export
Transcript exports support TXT, DOCX, PDF, VTT and SRT. Select the format your next tool accepts. Subtitle formats carry timing; a text document is more suitable for reading and editing prose.
Generated notes are separate from the transcript. If you need a shorter document, follow video summarization and check that output independently.
When something goes wrong
For missing text, check upload and transcription status before retrying. A transcription retry cannot restore sound absent from the recording. See transcript troubleshooting.
Review Accuracy and Methodology before relying on a transcript for a consequential decision or quotation.