This guide explains how to get details about the different meeting artifacts generated by a conference using the Google Meet REST API.
An artifact is a file or data generated by Google Meet in response to a conference . This includes videos and other artifacts, such as transcripts.
To produce artifacts, participants must generate them in Meet before the conference ends. Transcripts operate independently of recordings and you don't have to record the meeting to generate a transcript. For more information, see Record a video meeting and Use Transcripts with Google Meet .
If you're a meeting space owner or participant, you can call the get() 
and list() 
methods on the recordings 
, transcripts 
, and the transcripts.entries 
resources to retrieve artifacts.
Authenticating and authorizing with user credentials lets Google Meet apps access user data and perform operations on the authenticated user's behalf. Authenticating with domain-wide delegation lets you authorize an application's service account to access your users' data without requiring each user to give consent.
Artifact retention
After a conference ends, Meet saves recordings and transcripts to the meeting organizer's Google Drive. By default, Meet artifacts are retained according to Drive rules.
Transcript entries provided by the Meet REST API are deleted 30 days after the conference ends.
You also can manage the retention of Meet artifacts separately using Meet-specific retention rules in Google Vault. For more information, see Retain Google Meet data with Vault .
Recordings
The following sections detail how to get information about recordings in a conference record.
Meet has access to the recording ID after the recording file is generated. It's possible a user might delete the recording file from Google Drive, yet the unique name is still returned in Meet.
The  recordings 
 
resource includes both the  DriveDestination 
 
and the  State 
 
objects.
The DriveDestination 
object holds the export location in Drive
where the recording is saved as an MP4 file. To download the recording or to
play it back in a browser, use the object's exportUri 
field value. If you're
familiar with the Google Drive API, the file 
field value corresponds to the id 
in the  files 
 
resource. For
more information, see Download and export files 
.
The State 
object holds the status of the recording session. This lets you
query if the recording has started, if the recording has ended but the recording
file isn't ready, or if the recording file is generated and ready to download.
Get details about a recording
To get details about a specific recording, use the  get() 
 
method
on the  recordings 
 
resource with the name 
path parameter. If you don't know the recording name,
you can list all recording names 
using the list() 
method.
The method returns an instance of a recordings 
resource.
The following code sample shows how to retrieve a specific recording:
Java
Node.js
Python
Replace the recording name with the name of the specific recording ID in a conference record.
List all recordings
To list details about all recordings in a conference record, use the  list() 
 
method
on the  recordings 
 
resource with the parent 
path parameter. Format: conferenceRecords/{conferenceRecord} 
.
The method returns a list of conference recordings, ordered by startTime 
in
ascending order, as an instance of a recordings 
resource. To adjust the page
size, see Customize pagination 
.
The following code sample shows how to list all recordings in a conference record:
Java
Node.js
Python
Replace the parent value with the name of the conference record.
Transcripts
The following sections detail how to get information about conference transcripts in a conference record.
Meet has access to the transcript ID once the transcript data is generated. It's possible a user might delete the transcript file from Drive, yet the unique name is still returned in Meet.
The  transcripts 
 
resource includes both the  DocsDestination 
 
and the  State 
 
objects.
The DocsDestination 
object holds the export location in Drive
where the Google Docs transcript file is saved. To fetch the content or to
browse the transcript in a browser, use the object's exportUri 
field value. If
you're familiar with the Google Docs API, the document 
field value corresponds to
the documentId 
in the  documents 
 
resource.
The State 
object holds the status of the transcript session. This lets you
query if the transcript session has started, if the transcript session has ended
but the transcript file isn't ready, or if the transcript file is generated and
ready to download.
Get details about a transcript
To get details about a specific transcript, use the  get() 
 
method
on the  transcripts 
 
resource with the name 
path parameter. If you don't know the transcript name,
you can list all transcript names 
using the list() 
method.
The method returns an instance of a transcripts 
resource.
The following code sample shows how to retrieve a specific transcript:
Java
Node.js
Python
Replace the transcript name with the name of the specific transcript ID in a conference record.
List all transcripts
To list details about all transcripts in a conference record, use the  list() 
 
method on the  transcripts 
 
resource with the parent 
path parameter. Format: conferenceRecords/{conferenceRecord} 
.
The method returns a list of conference transcripts, ordered by startTime 
in
ascending order, as an instance of a transcripts 
resource. To adjust the page
size, see Customize pagination 
.
The following code sample shows how to list all transcripts in a conference record:
Java
Node.js
Python
Replace the parent value with the name of the conference record.
Transcript entries
The following sections detail how to get each participant's speech as transcript entries during a conference transcript session.
The transcript entry contains transcribed text of the participant's voice, up to
a maximum of 10,000 words. The spoken text languageCode 
(for example, en-US 
)
is also included as IETF BCP 47 syntax 
.
Each transcript entry is connected to a participant 
name in the conference. To
retrieve participant details, see Get details about a
participant 
.
Transcript entry data is available for 30 days after the conference ends.
Note that the transcript entries returned by the Meet REST API might not match the transcription found in the Docs transcript file. This occurs when the transcript file is modified after generation.
Get details about a transcript entry
To get details about a specific transcript entry, use the  get() 
 
method on the  transcripts.entries 
 
resource with the name 
path parameter. If you don't know the transcript entry
name, you can list all transcript names 
using the list() 
method.
The method returns an instance of a transcripts.entries 
resource.
The following code sample shows how to retrieve a specific transcript entry:
Java
Node.js
Python
Replace the transcript entry name with the name of the specific transcript entry ID in a transcript.
List all transcript entries
To list details about all transcript entries in a transcript, use the  list() 
 
method on the  transcripts.entries 
 
resource with the parent 
path parameter. Format: conferenceRecords/{conferenceRecord}/transcripts/{transcript} 
.
The method returns a list of structured transcript entries per conference
transcript, ordered by startTime 
in ascending order, as an instance of a transcripts.entries 
resource. To adjust the page size, see Customize
pagination 
.
The following code sample shows how to list all transcript entries in a transcript:
Java
Node.js
Python
Replace the parent value with the name of the conference record and the transcript name.
Customize pagination
Pass the following query parameters to customize pagination of recordings, transcripts, and transcript entries:
-  pageSize: The maximum number of items to return. The service might return fewer than this value. If unspecified, at most 10 items are returned. The maximum value is 100; values more than 100 are automatically changed to 100.
-  pageToken: A page token, received from a previous list call. Provide this token to retrieve the subsequent page.

