File details

example file details page

The controls on this page enable you to:

  1. Play the entirety of the audio file, use the slider, or navigate to specific moments by clicking on a section of the transcript.

  2. Download the file in audio MP3 format, and the transcript in text or JSON formats.

  3. Read the transcript, taking advantage of highlighted search terms.

  4. Select tabs to switch between viewing File Details including call metadata and emotion, Application Scores, and Tags.

When folders are created, folder options define what type of audio will be added to the folder and how the audio should be broken down in the transcript.

  • Transcripts of two-channel, stereo audio with one speaker on each channel are divided into "speaker turns" that show you who was speaking during each part of the transcript.

  • Transcripts of one-channel, mono audio files with a single speaker are broken up into "utterances" to make them easier to read and to simplify drilling down into a single utterance.

  • Transcripts of one-channel, mono audio files with two speakers may display speaker turns due to V‑Spark 's "diarization" feature. However, if diarization does not detect multiple speakers to trigger different speaker turns, the transcript will display all audio as a single speaker, broken up into utterances.

Highlighted areas within a transcript use different colors to indicate the following:

  • green - positive emotion identified

  • grey - both positive and negative emotional content

  • red - negative emotion identified

  • (none) - no detectable emotion

Bold words in the transcript identify those words or clauses that have emotional content.

Clicking on a section of the transcript will also allow you to listen to the audio at that point in the conversation. The audio player skips over sections of silence longer than 10 seconds when Skip Silence is turned on. In the previous example, Show Emotion is ON and Skip Silence is OFF.

For 2-channel audio calls that receive an overtalk score greater than 0%, the score is highlighted by blue text. When speaker overtalk occurs within a transcript, the affected speaker turns will display a solid blue border on the left-hand side so that they can be easily spotted and reviewed if necessary.

If you navigate to the File Details page after conducting a search, your search terms appear highlighted in yellow, and arrows on the right hand side enable you to navigate between instances of your search terms.

Silence time is visualized in the transcript as [SILENCE (length of silence in seconds)]. For example, [SILENCE 4] indicates 4 seconds of silence. Only silence lasting longer than 3 seconds will be marked. You can filter calls based on the percentage of silence time using the Call Silence Time slider in the Files View of the Dashboard.

UI controls

To see information about an audio file in the current folder, click its name in the file list to open a File Details screen:

  1. Clear search term highlighting from transcript display

  2. Use the up and down arrows to view instances of your search terms

  3. Play call audio

  4. Select MP3 to download an MP3 audio file of the call. Select JSON to download a JSON file of the call transcript. Select Text to download a text file of the call transcript.

  5. Select a piece of the transcript to play audio from that point

  6. Show emotion for a visual representation of sentiment analysis

  7. Skip silent portions of the call for faster review

  8. Select to display a URL that links to this point in the transcript. You can copy this URL to the clipboard for use elsewhere.

  9. The blue border on the left of the transcript indicates portions of the call with high overtalk