Home Luware Nimbus Usage of Nimbus Service Settings

AI Service Settings

AI-assisted User Guidance for your Service

Nimbus allows for AI-driven features that help improve your overall service performance. Scenarios and features involve:

Companion:
- AI-driven Transcription on customer-service user interactions and post-call Summarization functions.
- Automated Tags and Codes suggestions based on the context detected during the call.
Virtual Assistants: Live Captioning of spoken dialogue.
Virtual Users:
- AI-driven Transcription of customer-AI user interactions.
- For both inbound and outbound audio calls, including Call On Behalf and Scheduled Tasks.

✅Companion, Virtual Users and related Transcription features each require individual Tenant Administrator consent. This consent needs to be granted once in the Data Privacy Tenant Settings > Artificial Intelligence section by individually enabling the use of AI-driven features.

☝️Without the consent given, AI features described below will be inactive for all services. A warning will inform about

🔎DATA STORAGE SETTINGS

Note that there are data storage toggles in the features below.

While enabled (Default): Data is stored or 30 days¹. Power Automate Connector > Trigger Events and Flow Actions for Companion and Virtual User can be leveraged to retrieve the data.
While disabled: Data will be cached and (when optionally enabled) shown in the Nimbus UI for 1 hour¹. Caching is done to allow retrieval and storage of the transcription data via Nimbus Power Automate Connector.

¹🔎Any data not stored into external systems will be discarded after the retention period and will not show in historical listings anymore (e.g. Sessions List, My Sessions). Data storage retention settings apply equally to inbound and outbound calls. For more details on data storage and retention, refer to the Nimbus Security Whitepapers in the Documents section

Companion Features

Precondition: Speech Recognizer

✅ All Transcription or Live Captioning features described on this page requires prior setup of a Speech Recognizer in the Nimbus Configuration.

Speech recognizers are preconfigured Nimbus (Administration) items that point to a 3rd party service for voice processing. A Speech Recognizer is required to specify the language and engine used by any transcription features mentioned below.

Please use the tabs below to navigate:

Transcription

INC Transcription Preconditions

PRECONDITIONS

✅Related Admin Use Case: Refer to Use Case - Setting Up Transcription for detailed step-by-step instructions.

Audio / Video Modality: AI driven Transcription, Summarization, Live captioning, Codes & Tags suggestions are available for both inbound and outbound audio calls.

Nimbus service and user licensing

🔎Transcription features have service and user requirements. These requirements apply for either mid-session Live Caption or post-session Transcription / Summarization features.

Service requirements

Enterprise Routing Contact Center

License requirement: Only Enterprise Routing or Contact Center services have the AI tab and related Nimbus Features available.
Feature toggle: As Administrator/Team Owner you can enable single Companion features via AI Service Settings.
Data display:
- To show the transcription data to Nimbus users, you need to enable the “Companion” widget on the Extension Service Settings.
- Of course you can also opt to not show the transcription data, instead leveraging it via the Nimbus Power Automate Connector. → More on this below.

Service User requirements

Companion

✅ The service requirements needs to be met.

✅ Additional steps are as follows:

License requirement:
- Your tenant needs to have sufficient available Companion licenses assigned in Licenses Tenant Settings. You can review and bulk-assign available licenses via License Management.
- One individual license per Nimbus user (human agents) is required.
Feature toggle: This can be done by any Administrator in the General User Settings.
⮑ Once enabled, the My Sessions page will show a new “Companion” widget for this user, showing (individually) enabled features sorted into tabs.

Virtual User transcription

Virtual Users

✅ The service requirements needs to be met.

✅ Additional steps are as follows:

License requirement:
- Your tenant needs to have sufficient available Virtual User licenses assigned in Licenses Tenant Settings
- One individual license and setup per Virtual User is required. You can review and bulk-assign available licenses via License Management.
Feature toggle: Transcription is enabled in AI Service Settings.
⮑ Once enabled, any Virtual Users (AI agents) interacting with customers within that service's workflow can also be transcribed.
Data processing: As no human agent is taking the call, the transcription is not automatically shown in the Nimbus UI. The transcript must be handled via the Nimbus Power Automate Connector. → More on this below.

Speech Services (Speech Recognizer)

🔎Features described in the following use “Speech Services” (in Nimbus configured as Speech Recognizers) provided by 3rd party vendors. To offer our customers both convenience and flexibility, you may pick between a Nimbus-native implementation and Azure speech services. Differences are explained in the following:

INC Speech Recognizer service comparison

	Nimbus AI Services	Azure AI Services
Benefits	Nimbus native speech recognizer, available out-of--box Low initial setup and maintenance Centrally managed in the Nimbus UI	Full flexibility in configuration and LLM choice Possibility to train the model dependent on your use case and business needs
Challenges	Less flexible and no custom configuration possible Usage based on Luware “fair use” policy as part of your license.	Will cause monthly ACS costs outside of your Nimbus subscription (e.g. APIs for speech services).
Setup	Low barrier of entry, no API knowledge required Data data storage beyond retention time is done via Power Automate Connector.	Administrators need to set up Speech Services and manage the API. Read our Use Case - Setting Up a Speech Recognizer in Azure Portal and Nimbus for details on this. The API key needs to be added in the Nimbus Speech Recognizer settings. Data storage regions as specified by Microsoft AI services.

Power Automate Connector

🔎Conditional step: The Nimbus Power Automate Connector can extract Transcription/Summarization data from customer sessions. There are separate Trigger Events and Flow Actions for Virtual User (automated AI Agents) and Companion (while assisting human Nimbus users).

To retrieve and handle the data, the following steps need to be performed by an administrator

Transcription setup for ►	Companion (Human) Service Agents	Virtual Users (AI) Service Agents
Mandatory steps	None, Nimbus automatically shows Transcription data in the UI when enabled in service settings.	Set up a Power Automate flow that uses the “Virtual User” Trigger Events to react to transcription-relevant Nimbus session events. Follow up with “Virtual User” Flow Actions to capture the data for any further processing, e.g. as a message, stored in a CRM, etc.
Optional steps	Set up a Power Automate flow that uses the “Companion” Trigger Events to react to transcription-relevant Nimbus session events. Follow up with “Companion” Flow Actions to capture the data for any further processing, e.g. as a message, stored in a CRM, etc.	None. The Transcript must be handled within the session data retention time, otherwise it will be lost.
Related pages	Use Case - Setting Up Transcription Use Case - Analyzing a Transcript Use Case - Sending a summary and follow-up Tasks to an Agent	Use Case - Integrating Nimbus Virtual User with external systems Use Case - Analyzing a Transcript Use Case - Sending a summary and follow-up Tasks to an Agent

INC Azure Billing Transcription

AZURE BILLING

The usage of the Transcription feature will cause additional monthly ACS costs. The costs are determined by Microsoft. Also see https://azure.microsoft.com/en-us/pricing/details/cognitive-services/speech-services/.

Before enabling the Transcription feature, get in touch with your Luware Customer Success specialist to discuss terms, usage scenarios and necessary setup procedures.
Please note that Nimbus and Transcription support does not cover pricing discussions or make recommendations based on Microsoft Azure infrastructure.

When active, the Transcription feature transcribes call contents between customer and Nimbus users (Agents) using Voice-to-Text capabilities. It improves quality assurance and increases the efficiency of call processing, for example making it easier to summarize past customer interactions and possibly transfer them to a CRM system for easier search and later analysis. Transcription is available for both inbound and outbound audio calls.

Once a call has ended, a Voice Transcription is generated in the background, by leveraging the Nimbus Power Automate Connector > "Companion" Trigger Events and Flow Actions. You can then forward the generated transcription to any target, e.g. by wrapping it into an Adaptive Card as Teams chat message or via Email.

Optionally you can also show the transcription in My Sessions, allowing users to have a direct written record of previous conversations with customers. Transcriptions for concluded outbound calls are also available in the historical sessions view in My Sessions.

Example of a finalized Transcription on the "My Sessions" page

Summarization

✅ PRECONDITIONS

Companion - Summarization is an optional feature as part of Transcription:

The Transcription feature must be enabled on the Companion Service Settings, as the Summarization is generated from it.
Companion Service Settings also steer the individual feature availability to users, as well as the data storage retention time.
Each user accessing the feature requires a Companion License, applied via General User Settings.

The Nimbus Companion integrates AI summarization as part of a broader suite of Transcription features, aimed at improving service performance and agent efficiency. During a call - when Live Captioning is enabled - Nimbus transcribes the conversation in real time. After the call, the transcript is processed by an AI service to generate a summary with the following aspects:

Aspect	Description
Title	A title generated from Call Direction, Service, and User that accepted the task.
Issue	Shortened problem description by the Customer.
Resolution	Summary of how the issue was resolved.
Recap	A brief paragraph summarizing the interaction.
Narrative	Detailed call notes or chat summary of the entire Customer/User Interaction.

Summarization Outputs

Via My Sessions: Summarization outputs can be presented to the agent via My Sessions > Companion widget.
💡All aspects shown within the Companion Widget > Summarization can be hovered over and copied.
✅ Precondition: The “Companion” widget itself needs to be enabled in the Extensions Service Settings in order to be shown to users.

Via Nimbus Power Automate Connector: Administrators also have the possibility to flexibly retrieve and store Summarization outputs using Companion Flow Actions and the User/Service SessionID. This can be useful to keep a record of calls, e.g. within a CRM or long-term data storage and evaluation platform.

Known Limitations

Summarization relies on the Transcription feature and therefore shares the same design limitations.

INC Transcription Limitations

☝️ KNOWN TRANSCRIPTION & SUMMARIZATION LIMITATIONS

💡We are actively working on further improvements to the following limitations:

Transcription data is not part of service/user transfers. Data is kept within the current customer/user session. This capability will be provided with an upcoming release.
Virtual User Reporting: Virtual User sessions and their transcripts will be included in Virtual User session reporting as part of a separate, upcoming capability. This is not available yet.
Power Automate Connector: New “Virtual User” data fields in the Trigger Events / Flow Action may still be erroneously labeled as “Companion”. The terminology will be corrected in a separate update. The Virtual User / Companion connectors separate their data as intended.

Troubleshooting FAQ

INC Transcription Troubleshooting

Not seeing any transcripts in the Transcript widget can have several causes. The following table lists error messages and explains why they are shown.

Message shown	🤔 Why do I see the message?
No transcription is available for this interaction.	There was no transcription generated for the selected conversation. This could happen … … when there are issues with the configured Speech Recognizers. … when there were issues with the microphone/speakers. … or simply when no one said anything during the transcription.
No transcription is available for this interaction OR Task was not accepted.	The selected call was not accepted and therefore there is no transcribed conversation.
Transcription not available for this session. No consent granted.	The customer explicitly declined transcription consent before being connected to a Nimbus user. The transcription pipeline was skipped for the entire session. No transcription, summarization, or AI-powered suggestions are available. 💡This message only appears when the customer explicitly declined consent. If the Record Consent activity is not included in the workflow, transcription proceeds normally.
User Transcription License is missing.	The user doesn't have a Companion license assigned. → Licenses can be assigned to users in Users > General User Settings > Licenses. You can also inspect your available license contingency via Admin Portal > License Management.

🔎 BY DESIGN

💡The following points are not limitations, but important to note:

Transcription Feature availability:
- Transcription is a prerequisite to Live Captioning, Summarization, AI suggested Codes & Tags
- Not for Copilot / Direct Line 3.0 Virtual Users. Transcription features apply Virtual Users built on GPT Realtime bots only.
- Transcription relies on external services and APIs. When the feature is disabled or unavailable (e.g. throttling or service impediments) info messages are shown in the frontend UI widget. Nimbus task handling and user-customer interactions themselves are not affected by this and will continue normally.
Virtual Users:
- Not for Copilot / Direct Line 3.0. Transcription features apply Virtual Users built on GPT Realtime bots only.
- Post-session only. The transcript is produced after the session ends; there is no live/real-time transcription during a call.
- No (default) My Sessions view. Virtual User transcriptions do not appear on the Nimbus UI — they are consumed through Power Automate.
- No audio storage. Only the generated transcription data is processed in Power Automate. Audio recordings are discarded after automated processing.
Transcription scope:
- 3rd-party participants: Third parties, consultants and call conference attendees are not part of the transcribed content. Only the transcription data between Nimbus users / AI agents and the calling customer is kept.
- Supervision - As long as Supervisors remain in Listen or Whisper mode during a conversation, their voice is not being transcribed. Only during ”Barge In" they are part of the conversation and transcription is active.
- Summarization items generated from the Transcription might be missing when there is not enough data to draw from.

Codes Suggestion

✅ Precondition: Transcription is required to be enabled and a Transcriber set in order to use Codes Suggestion.

The Codes Suggestion feature provides you suggestions in My Sessions after a call is finished. If you haven't added codes to the session yet, Companion automatically suggests them. In case you have already added your codes, you need to click on the suggest icon to see and apply code suggestions. Code suggestions are also available for concluded outbound calls in the historical sessions view.

AI-suggested Codes in My Sessions, based on what the AI matched with the Transcription

💡When configuring Codes in your Configuration, you can add the context related to this code in the field to support Nimbus Companion in suggesting codes.

Tags Suggestion

✅ Precondition: Transcription is required to be enabled and a Transcriber set in order to use Tags Suggestion.

The Tags Suggestion feature provides you up to 5 tag suggestions in My Sessions from the list of already assigned tags (to service sessions). After a call is finished and you haven't added tags to the session yet, tags suggestions are shown. You can either select specific tags or apply them all. If you have already selected your own tags before, they don't get overwritten – in this case, you need to click on the suggest icon to see and apply tags suggestions. Tag suggestions are also available for concluded outbound calls in the historical sessions view.

Companion Follow-Up Actions

For Service Owners

✅Enabling the feature visibility:

For Companion Transcription and Summarization features to be visible to your users, the "Companion" widget must be enabled via Extensions Service Settings > Widgets. Of course you can opt to not show these features, while Companion features still run in the background.
💡Workflows can be set up for customers to opt out of transcription via the Record Consent workflow activity. When a customer opts out, transcription will not be generated for that session..
💡Please note that Transcription-related features only apply for new sessions, not retroactively on past sessions.
Once enabled, you can inform your Service team, so they can pay attention in their My Sessions view accordingly.

For Administrators

✅Power Automate alternative for Companion data: You can also leverage the Nimbus Power Automate Connector by using "Companion" related Trigger Events and Flow Actions, as described in our example Use Case - Analyzing a Transcript. While adding complexity in configuration, the advantage is that you have more fine-tuned control over the transcription data, allowing you to process or store it externally, or opt to showcase it to additional users via the use of Adaptive Cards. The Power Automate triggers and actions also fire for outbound conversations.

Virtual Assistants

Live Captioning

✅ Precondition: Transcription is required to be enabled in order to use Live Caption.

INC Public Preview Beta Feature

This feature is in PREVIEW and may not yet be available to all customers. Functionality, scope and design may change considerably.

The Live Caption feature converts ongoing call participant voices directly into text, using Voice-to-Text capabilities. Contents are displayed live on the My Sessions page. Live Captioning is also available for outbound audio calls.

Example of ongoing Live Captioning on the "My Sessions" page

Virtual Users