1. Click  Menu > Analytics > Analytics Workspace.
2. Click Add () and select Interactions.
3. Click the row of the interaction whose transcript you want to view.
4. Click the Transcript tab.

   Note: As the interaction recording plays, the corresponding words (up to three), are highlighted with a blue background in the transcript. After the words are spoken, they are grayed out in the transcript. For more information, see Work with an interaction overview.

   Click the image to enlarge.

   

1. On the Transcript tab, in the Search field at the top of the transcript, enter the words that you want to find.
2. Press Enter. 

The words and sentences in the transcript are divided into speaker sections: customer, IVR, ACD, or agent.

On the left side of the screen, a time stamp distinguishes every section. This time stamp indicates when the speaker began to speak and two of the following icons representing the speaker.

Click the image to enlarge.

To protect the privacy of your customers, automatic redaction is enabled by default to ensure that any sensitive PII or PCI data collected from an interaction is automatically redacted from voice transcripts. For example, instead of a credit card number the transcript includes [card number]. When PII or PCI is data is detected and redacted, the user hears silence in the audio playback where the data was found.

This list is a complete list of redacted data:

[card number 
[card expiry date]
[national id]
[postal code]
[sensitive data]
[phone]
[name]
[location]
[user info]
[email] 

A select number of users can access transcripts that are not redacted only if they have the Recording > Recording – View Sensitive Data permission. By default, roles do not come with this permission. An administrator manually grants the permission.

For more information, see Enable automatic redaction of sensitive information.

1. Click  Menu > Analytics > Analytics Workspace.
2. Click Add () and select Interactions.
3. Click the row of the interaction whose transcript you want to view.
4. Click the Transcript tab.
5. Click Copy Transcript  in the upper right corner of the transcript. 

The following metadata is copied.

• Interaction Type
• Interaction ID
• Message Type (only available for a message interaction)
• Start Time (User’s timezone)
• End Time (User’s timezone)
• Duration
• Internal Participants
• External Participants
• Transcript:
  • Date/Time (User’s timezone)
  • Participant Type (Internal/External)
  • Participant
  • Text
• Interaction Summary

• Genesys Cloud AI Experience tokens metering and pricing
• If a user clicks the button to translate an interaction into Spanish, closes the interaction, and then reopens it later to translate into Spanish again, is that counted as one translation or two?
• If another user opens the same interaction and translates it to Portuguese instead of Spanish, is that counted as one translation or two?

Key points:

• Availability
  • The translation option is available on the Interaction Details page when a transcription is present.
• Types of interactions supported
  • Voice transcriptions, chats, digital/web messages, and email interactions (body only) can be translated.
    Interactions with an HTML toggle enabled are not supported.
• User interface elements
  • A translation icon (which changes appearance when active) is displayed.
  • A label at the top of the transcript panel indicates that translation is in progress and shows the source and target dialects.
  • You can select a different target language via a drop-down menu, where languages are listed in their native scripts (for example, Hebrew appears as עברית).
• Considerations
  • For voice interactions, only approximately the first 25 minutes (or 10,000 characters) are translated.
  • For digital interactions, only the first 10,000 characters are translated.
  • The feature is limited to 1,000 translations per organization per day.
  • During translation, original text features like topic tags, search functionality, karaoke, sentiment markers, and redactions behave differently. For example, copying the transcript always copies the original text.
  • Any redacted text in the original transcript is not translated. Note: Sometimes, if the translation service handles it that way, the redacted entity name appears in brackets.
  • After initial translation, the translated text is cached for up to 20 hours and made visible after activating the translation icon.
    

Work with on-demand transcription translation

1. Navigate to the interaction that you want to review. Ensure that the interaction contains a transcription (voice, chat, digital message, or email body).
2. Locate the translation icon . Within the transcript panel, look for the translation icon. This icon is visible only when a transcription is present.
3. Click or toggle the translation icon. Once activated, the icon changes its appearance to indicate that the translation process has started. A label appears at the top of the transcript panel, stating that the text is being translated from the source dialect to your selected target dialect.
4. Wait for the translation to complete. A spinning icon appears, usually on the upper right of the interaction page, while the system processes the translation. For voice interactions, expect the translation to appear within 4 seconds for at least half of the requests, and within 10 seconds for nearly all requests. Email translations can take longer.
5. Review the translated transcript. Once the translation is complete, the translated text appears underneath each line of the original transcript.
   Some elements (for example, topic tags, sentiment markers, and search functionality) remain based on the original text.
6. Adjust the Target Language (if necessary). If the automatically selected target language is not your preferred dialect, click the drop-down menu in the translation area and choose your required language. Languages in the drop-down appear in their native format (for example, Hebrew as עברית).

Disable translation

To revert to the original transcript, just toggle off the translation option. The translation label disappears, the icon returns to its original state, and the transcript shows the original language.

• Quality assurance
• Coaching and training
• Identifying trends in customer interactions

Agent-level summaries provide AI-generated insights into the actions of both human and virtual agents during customer interactions. To quickly review agent performance and make evaluations faster and more meaningful, supervisors can use these summaries. The feature is available when Agent Copilot is enabled and applies across all interaction types, giving supervisors a complete view of customer engagements. Both the AI-generated summary and any edits made by the agent are displayed, offering a transparent and accurate record of activity.

An AI-generated outline organizes an interaction into chapters, making it easier to review high-level conversation flow and quickly navigate to specific parts, creating a clearer path to insights. Each section includes a generated header, a brief description, and the related transcript content. The outline preserves the chronological order of the interaction and includes every transcript sentence exactly once.

AI-generated outlines are generated at the communication level. After the outline is generated, it is available as long as the transcript exists.

To view the AI-generated outline:

1. Click  Menu > Analytics > Analytics Workspace.
2. In the Default section, search for Interactions, and then click the view name to open it.
3. Click the row of the interaction whose transcript you want to view.
4. Click the Transcript tab and select an interaction. By default, the transcript opens in the regular transcript view.
5. Turn on the AI outline or Collapse all outline segments .
6. To return to the standard transcript, turn off the AI outline toggle.

• AI summaries are available for voice, chat, and messaging transcripts (email interactions are not summarized).
• Data retention
  • The AI summary follows the same retention policy as the recording and transcript.
  • If a recording is deleted, the AI summary is also deleted.
• Transcript modifications
  • If new transcripts are added or some are deleted from an interaction, the summary is regenerated.
  • If all transcripts in an interaction are deleted, the summary and AI Insights are permanently removed.