Transcription (beta)

This app is in "beta mode". This means that currently it can be inconsistent in its performance or results. While we continue to improve this feature, we consider it to be experimental in its current form.

The Transcription application processes audio and video files to generate precise text outputs. It supports efficient conversion of recorded content into text, providing a reliable solution for creating transcribed text records of meetings, presentations, and conversations.

In this article:

Transcribing an audio or video file
Working with transcripts
Diarisation: Managing participants in a transcript
Supported languages

Transcribing an audio or video file

To generate a transcription of an audio or video file:

Navigate to the Transcription application in the sidebar.
All previous transcripts are listed on the Transcription page.
Click Upload new.
The Upload file popup appears:
Select a file in one of two ways:
1. Press Click to select in the popup, select a file from your system’s file selector.
2. Drag and drop a file into the popup area.
Click Upload in the popup.

A new card appears in the Transcription page representing the uploaded file, and the transcription process starts immediately:

PhariaAssistant Transcription - transcription in progress

The transcription process can take a few minutes. When it is done, you receive a notification, and the card progress is updated:

PhariaAssistant Transcription - transcription complete

Limitations

You can upload only one document at a time.
A newly uploaded document replaces the previously uploaded document.
Transcribe supports .mp3, .mp4, .m4a, .wav, .mpga, .mpeg, and .webm file formats.
The maximum file size is 200 MB.

Working with transcripts

The Transcriptions page displays all your completed transcripts. You can search for specific transcripts by entering text in the search bar at the top of the Transcriptions page. The transcript display is filtered to show only those whose names include the search text.

You view a transcript by clicking on its card in the list. Each transcription looks similar to the following:

PhariaAssistant Transcription - display transcription

You can do the following:

Edit the transcript text.
Copy the transcript text.
Export the transcript as a Word file (.docx).
Rename and delete the transcript.

See also identifying and assigning (diarisation) participants in the transcript.

Editing a transcript

You can edit the transcript text directly in the Transcript tab. At the top right of the transcript, click the edit icon and edit the text as desired. Click Save to register your changes.

Copy a transcript

You can copy the text of a transcription by clicking the copy icon . The transcript text, including the timestamps, is copied to the buffer, and you can paste it elsewhere.

You can also copy an individual section of the transcript by clicking the copy icon that appears in the top right of a section when you hover your mouse in it.

Exporting a transcript as a Word file

You can export a transcript to a Word (.docx) file on the Transcription page. In the card of the transcript you want to export, click the kebab menu icon in the top right and select the download icon .

Alternatively, you can open the transcript you want to export, and click the download icon at the top right.

Renaming a transcript

When you upload a file to transcribe, the transcription takes the name of the uploaded file. You can change this name on the Transcription page. In the card of the transcript you want to rename, click the kebab menu icon in the top right and select Rename file.

You can also do this directly in the transcript page.

Deleting a transcript

You can delete a transcript from the Transcription page. In the card of the transcript you want to delete, click the kebab menu icon in the top right and select Delete.

You can also do this directly in the transcript page.

Diarisation: Managing participants in a transcript

You can identify the speakers in an audio or video recording to indicate the correct participants in the transcript.

Adding and removing participants

In the Participants tab, the currently identified participants are listed:

You can manage these as follows:

To hear the assigned participant’s voice to confirm that the identification is correct, click the play icon next to their name, listen, and click to stop the playback.
To edit the name of a participant who is identified in the recording, click the edit icon next to their name, edit the name, and click Save.
To add a new participant, click Add participant, enter the participant’s name in the field, and click Add.
To remove a participant from the list of participants, click the trash icon next to their name.

Assign a participant in the transcript

In the Transcript tab, click the name of the currently assigned participant for a text. Select a new name from the Reassign participants popup to indicate that the text was spoken by that participant:

Supported languages

The Transcription application uses the Whisper model for transcriptions. The official list of supported languages can be found here.

The Transcription application supports the following languages:

Afrikaans
Arabic
Armenian
Azerbaijani

Belarusian
Bosnian
Bulgarian

Catalan
Chinese
Croatian
Czech

Danish
Dutch

English
Estonian

Finnish
French

Galician
German
Greek

Hebrew
Hindi
Hungarian

Icelandic
Indonesian
Italian

Japanese

Kannada
Kazakh
Korean

Latvian
Lithuanian

Macedonian
Malay
Maori
Marathi

Nepali
Norwegian

Persian
Polish
Portuguese

Romanian
Russian

Serbian
Slovak
Slovenian
Spanish
Swahili
Swedish

Tagalog
Tamil
Thai
Turkish

Ukrainian
Urdu

Vietnamese

Welsh