AWS Transcribe
Collect AWS Transcribe speech transcription monitoring data.
Configuration¶
Install Func¶
We recommend enabling Hosted Func under Guance Integrations - Extensions. All prerequisites are installed automatically; continue with script installation.
For a self-hosted Func deployment, see Self-host Func.
Activate Script¶
Note: Prepare an AWS AK that meets the requirements. For simplicity, you can grant the global read-only permission
ReadOnlyAccessdirectly.
Activate the script in Hosted Func¶
- Log in to the Guance console
- Click the 【Integration】 menu and select 【Cloud Account Management】
- Click 【Add Cloud Account】, select 【AWS】, and fill in the required information on the interface. If you have already configured the cloud account information before, ignore this step
- Click 【Test】, and if the test is successful, click 【Save】. If the test fails, please check if the relevant configuration information is correct and test again
- Click 【Cloud Account Management】, and you can see the added cloud account in the list. Click the corresponding cloud account to enter the details page
- Click [Integrations] on the cloud account details page. Find
AWS Transcribein theNot installedlist and click [Install] to open the installation dialog. Select the transcription regions in use and the metrics to collect, then install. Region and metric support differs between batch and streaming transcription; choose according to the service in use.
Activate the script manually¶
- Sign in to the Func console, click [Script Market], open the Guance script market, and search for
integration_aws_transcribe. - Click 【Install】 and enter AWS AK ID, AK Secret, and the account name.
- Click [Deploy Startup Script]. The system automatically creates the
Startupscript set and configures the corresponding startup script. Replace theregionsplaceholder in the startup script with the actual Transcribe regions, separated by ASCII commas, and setcloudwatch_configs['targets'][0]['metrics']to['LimitExceededCount', 'SuccessfulRequestCount', 'SyncServerErrorCount', 'SyncUserErrorCount', 'ThrottledCount', 'TotalRequestCount']. - after enabling it, open"Management / Automatic Trigger Configuration"to view the automatic trigger configuration.Click[Run],to run it immediately,without waiting for the scheduled interval.wait a moment,you can view execution records and related logs
Verification¶
- Under 「Management / Automatic Trigger Configuration」, confirm that the task is configured and review its records and logs for permission or API errors.
- In Guance, open [Metrics] and check the
aws_transcribemeasurement for the monitoring data.
Metrics¶
After configuring AWS cloud monitoring, the default metrics are listed below. You can configure additional metrics for collection.
AWS Transcribe metrics are in the aws_transcribe measurement.
| MetricName | MetricDescribe | Dimensions | Unit |
|---|---|---|---|
limit_exceeded_count_sum |
Requests exceeding quotas other than request rate, such as concurrent jobs, transcription count, or audio size (batch and streaming) | domain, service_type, operation, region_id | count |
successful_request_count_sum |
Successful batch transcription requests with response codes 200–299 | domain, service_type, operation, region_id | count |
sync_server_error_count_sum |
Batch transcription requests with server errors and response codes 500–599 | domain, service_type, operation, region_id | count |
sync_user_error_count_sum |
Batch transcription requests with user errors and response codes 400–499 | domain, service_type, operation, region_id | count |
throttled_count_sum |
Requests throttled for exceeding the requests-per-second quota (batch and streaming) | domain, service_type, operation, region_id | count |
total_request_count_sum |
Total successful and failed requests (batch and streaming) | domain, service_type, operation, language_code, region_id | count |