Session Transcription

Session Transcription allows you to produce a transcript directly from the live session, without the need to record it. Full transcript is available right after the session is finished.

Session Transcription for Grow (custom) plan customers is currently available on demand. Email us at embedded@whereby.com if you want to start using Session Transcription. Terms and conditions apply

Session Transcription is a supplementary feature of our paid Whereby Embedded plans. You can review the pricing and options on our site.

Session transcripts are created by live streaming Whereby session audio in real time. After the session is finished they are saved as text files accessible from the dashboard or via the API. You can use the transcripts as a standalone resource (eg. for compliance purposes) or send to an external service for post processing (eg. to derive key topics or create a session summary).

Setup

You can enable and configure Session Transcription globally for your account, or individually for each room. All transcripts can be downloaded manually through the dashboard, or programmatically with the API requests.

Global configuration

When you enable Session Transcription globally through the dashboard, these settings become the default for all rooms and sessions. Enabling Session Transcription globally will result in all sessions being transcribed, including sessions in rooms created previously. You can override these global settings by specifying the transcription on a per room basis

If you want to use Session Transcription for all your sessions, you can enable it globally for your account. Go to “Configure” → “Transcription” section of your customer portal and choose "Session Transcription" option. Then choose the trigger and the main language of your sessions.

You can choose between the following transcription triggers:

  • Manual The host will start and stop the transcription manually.

  • Auto-start (1 person) Transcription will start when the first person joins and end when the last person leaves.

  • Auto-start (2 people) Transcription will start when 2 people join a room and end when the last person leaves.

Per room configuration

If you want to use Session Transcription for some of your sessions, or if you need a different configuration for some of the sessions, you can configure Session Transcription individually for the room. Room parameters will override the global Session Transcription settings.

To do so, create the room with POST /meetings request and specify the transcription options with the "startTrigger" and language of your choice:

"liveTranscription": { 
    "language": "en", 
    "startTrigger": "automatic" 
    },

You can choose between "manual", "automatic" or "automatic-2nd-participant" triggers, and below you will find the list of supported languages.

It is not possible to combine multiple transcription triggers. If you choose one of the automatic triggers, the host will not be able to stop the transcription during the session.

When the session is transcribed, the participants see a notification circle in the top-left meeting status bar:

Supported languages

Session Transcription generates a transcript in the specified language. You need to declare the language used by your session participants in advance - in the global configuration or per room with POST /meetings request. After the room is created, you cannot change the language of the Session Transcription.

Session Transcription supported languages
  • Bulgarian (bg)

  • Catalan (ca)

  • Chinese (Mandarin, Simplified) (zh)

  • Chinese (Mandarin, Traditional) (zh-TW)

  • Czech (cs)

  • Danish (da)

  • Dutch (nl)

  • English (en)

  • Estonian (et)

  • Finnish (fi)

  • Flemish (nl-BE)

  • French (fr)

  • German (de)

  • Greek (el)

  • Hindi (hi)

  • Hungarian (hu)

  • Indonesian (id)

  • Italian (it)

  • Japanese (ja)

  • Korean (ko)

  • Latvian (lv)

  • Lithuanian (lt)

  • Malay (ms)

  • Norwegian (no)

  • Polish (pl)

  • Portuguese (pt)

  • Brazilian Portuguese (pt-BR)

  • Romanian (ro)

  • Russian (ru)

  • Slovak (sk)

  • Spanish (es)

  • Swedish (sv)

  • Thai (th)

  • Turkish (tr)

  • Ukrainian (uk)

  • Vietnamese (vi)

Download and delete transcripts

Transcripts are saved in Whereby-provided storage and are available for download soon after the session is finished.

Download the transcript manually from the Configure->Transcriptions section of your dashboard.

Transcripts are downloaded as .md files. From the Transcriptiosn section you can also create a session summary or delete the transcript.

You can to automate your transcription process programmatically with a combination of API requests and webhook events:

  1. When a transcript is ready, Whereby sends a transcription.finished webhook event. Hook onto that event to fetch the transcriptionId of the session that you want to transcribe.

  2. Using the transcriptionId, send a GET request to retrieve the download link of the transcription file. Transcripts are downloaded as .md files.

  3. All transcripts will be stored in the Whereby-provided storage until you delete them. If you want to minimize the time when your sessions' content is stored in the Whereby-provided storage, you can delete the transcript with a DELETE request.

Known limitations

  1. Session Transcriptions are not compatible with Breakout Groups feature. When using Breakout Groups, the transcript will cover the conversation from the main room, but the audio from individual groups will not be transcribed.

  2. Session Transcriptions are currently not considered to be HIPAA compliant. Avoid using Session Transcriptions in order to maintain HIPAA compliance of your Whereby sessions. Learn more about Whereby HIPAA compliant setup.

  3. Session Transcription is available for sessions up to 12 hours long.

Coming soon...

We’re excited about the future of API-assisted content processing and wanted to give you a sneak peek at what’s on the horizon. Upcoming features and improvements we’re actively working on:

  • <whereby-embed> methods to start and stop transcribing programatically.

  • Ability to save the transcript into customer-managed AWS S3 bucket.

  • Live preview of the transcript, visible to all session participants.

  • Ability to download the transcript by the host or participants.

  • Integration point to plug into the live transcript in real-time (eg. to send it into 3rd party processing tool like a chatbot).

Last updated