Batch

Available in Classic and VPC

You can create batches to process multiple speech recognition tasks at once. Once created, a batch automatically runs after a certain period, recognizing audio from media files located in the specified path, generating recognition result files and execution result files.

The Batch menu allows you to create and manage batches.

Batch interface

The basic description of the Batch menu for using CLOVA Speech is as follows.

View batch list

To view the information of created batches:

1. From the NAVER Cloud Platform console, navigate to > Services > AI Services > CLOVA Speech.
2. Click Batch.
3. When the batch list appears, check the batch information.
   • Batch ID: Auto-generated batch ID
   • Batch name: Name entered when creating the batch
   • Domain: Domain name / domain code associated with the batch
   • File format: Output file format generated by the batch
   • Input file storage path: Object Storage path where input files are stored
   • Result file storage path: Object Storage path for recognition results and execution result files
   • Status: Indicates whether the batch is available
     • Normal: Available
     • Error: Unavailable
       • Hover over to see the error cause
   • [Settings] button: Edit batch settings (See Edit batch settings)

Create a batch

To create a batch:

1. From the NAVER Cloud Platform console, navigate to > Services > AI Services > CLOVA Speech.
2. Click Batch.
3. Click [Create batch] button.
4. Enter the batch information:
   • Batch name: Enter the batch name
   • Domain ID: Click the input field and select the domain from the list
     • You can search by name or code
     • Free domains and domains already linked to a batch are not shown
   • File format: Select the output file format
   • Language settings: Choose the language for recognition
   • Speaker separation: Enable or disable speaker separation
   • Keyword boosting: Enable keyword boosting (See Short-utterance recognition builder Keyword Boosting, Long-form recognition builder Keyword Boosting for details)
5. Enter the Object Storage paths where input and result files will be stored.
   • Input file storage path: Click the field and select the Object Storage path for input files
   • Result file storage path: Select the Object Storage path where recognition result and execution result files will be stored
6. If necessary, click under Advanced settings configure Webhook settings.
   • URL: Enter the URL to receive batch results
   Note

   When each task is completed, a webhook containing recognition results is sent to the specified URL.

7. Click [Create] button.

• It may take some time for the batch to be fully operational.
• Recognition result files and execution result files can be found in the result file storage path.
• Files that have been recognized are automatically moved into a folder created under the input path using the batch name.

Change batch settings

To modify the settings of a batch you created:

1. From the NAVER Cloud Platform console, navigate to > Services > AI Services > CLOVA Speech.
2. Click Batch.
3. In the batch list, click [Settings] button for the batch you want to modify.
4. Modify the batch settings.
   • For details on each information field, refer to Create a batch.
5. Click [Edit] button.

Delete a batch.

To delete a batch you created:

1. From the NAVER Cloud Platform console, navigate to > Services > AI Services > CLOVA Speech.
2. Click Batch.
3. In the batch list, select the batch you want to delete, then click [Delete a Batch]. button.
4. When the confirmation popup appears, click [Delete] button.