V-Spark Online Help

Create a Folder

Folders may be created using the UI, or with the API using the /config endpoint. Before using either method, V‑Spark must have at least one company and one organization configured for ASR processing. Refer to Company and Organization Management for information about configuring companies and organizations.

Create a Folder with the UI

  1. Click fa-gear.svg Settings > Folders to navigate to the Monitored Folders screen.

  2. Click fa-plus.svg Add. The Create New Folder dialog displays.

  3. Specify the settings displayed on the first page of the dialog, as in the following example:

    createFolder-1-co.png

    Important

    Settings configured on the first page of the Create New Folder dialog can't be changed after the folder is created, except for options in the Servers dropdown. Processing audio with different settings requires creating a new folder.

    Use the first page of the Create New Folder dialog to configure the following options:

    1. Organization — Select the folder's company and organization. A folder's company and organization determine the availability of some of its ASR configuration settings.

    2. Folder Name — Enter a name for the new folder. Folder names are case sensitive, and must not be longer than 100 characters.

    3. Servers — Select the servers to be used for ASR processing. Multiple servers may be selected from the dropdown if available. Server options are based on the folder's company and organization configurations.

      Note

      Configuring a folder to use multiple ASR servers that do not use the same language models may result in an error message indicating that "One or more servers does not have this model anymore." This error will only impact ASR processing if the language model has been removed from all of the folder's servers.

    4. # of speakers — Select the number of speakers in the audio, either 1 or 2.

    5. Audio Type — Select either mono (single-channel) or stereo (two-channel) audio. Use results from the Audio Evaluator to verify the audio's properties.

      Note

      If a folder is configured with the # of speakers field set to 2 and the Audio Type field set to Mono, that folder's processed audio is separated, or diarized, into 2 channels. Each speaker is assigned to a different channel. Speakers in the resultant transcript are classified as either Agent or Client using side classification.

      To prevent diarization of audio with more than two speakers, configure the folder for Mono audio with 1 speaker.

    6. PCI Redact TextOptional. Select to enable redaction for text transcripts processed by this folder. Enabling text redaction causes the ASR engine to remove payment card information (PCI) and other sensitive numbers from text transcripts.

    7. PCI Redact AudioOptional. Select to enable redaction for audio processed by this folder. Enabling audio redaction causes the ASR engine to remove PCI and other sensitive numbers from the processed audio file.

  4. Click Next. The dialog displays additional folder configuration options, as in the following example:

    create-newFolder-2c.png

    Use the second page of the Create New Folder dialog to configure the following options:

    1. Link to Applications — Select the applications to be used for analysis on the transcription output for all audio processed by the folder. Refer to Linking and Unlinking Folders and Applications for more information.

    2. Model — Select the language models to be used for audio transcription requested by the folder. One model must be selected for each audio channel, and channels may use either identical or distinct language models.

      Model selection greatly impacts transcription accuracy because they are tuned to specific audio scenarios and environments, like voicemail or call centers. Refer to the language models documentation for more information.

    3. Which channel is the agent on? — Select the audio channel for agent speech. This option is used only with stereo audio folder configuration. A value of 0 refers to the left audio channel, and a value of 1 refers to the right audio channel.

      Important

      The agent channel value must be set accurately for transcription analysis, and it cannot be changed after the folder is created.

    4. Advanced settings — Enable and configure settings for callbacks, custom metadata, ASR options, and duplicate file rejection.

      Enabling Reject duplicate records by filename prevents the folder from accepting any file with the same name as one that has already been uploaded. The remaining advanced settings must be configured after being enabled.

      To configure advanced settings, enable them and click Next. An additional Create New Folder dialog displays for each enabled setting except for duplicate file rejection. The last page of the configuration dialog has a Create button.

      Refer to the following topics for more information about configuring other advanced settings:

  5. Click Create to close the configuration dialog and create the folder.

  6. View the ASR server status field for the newly-created folder on the Monitored Folders screen. A green circle fa-circle-green.svg in the server status field indicates an active and authenticated connection which means the folder is ready to process audio.