- Print
- PDF
Article summary
Did you find this summary helpful?
Thank you for your feedback
Breaking changes
- Analysis Worker service is now handling the text correction. Text normalization URL should be imported to Knovvu ML and its endpoint should be defined within the inventory.
- Knovvu ML should be upgraded to 1.52 for clustering and normalization services.
- PUT calls/{id} method has been changed into PUT calls/ {externalid}
- PUT Departments/{id} method has been changed into PUT Departments/ {externalid}
- PUT User/active-passive/{id} method has been changed into PUT User/active-passive/ {externalid}
- PUT User/{id}/profile-picture method has been changed into PUT User/{externalid}/profile-picture
- PUT User/{id} method has been changed into PUT User/{externalid}
- GET User/{id} method has been changed into GET User/{externalid}
- GET Conversations/{id} method has been changed into GET Conversations/ {externalid}
- GET calls/{id}/download method has been changed into GET calls/{externalid}/download
- PUT video-calls/{id}/video method has been changed into PUT video-calls/{externalId}/video
- PUT conversations/{id} method has been changed into PUT conversations/{externalId}
- PUT conversations/{id}/screen-record method has been changed into PUT conversations/{externalid}/screen-record
- PUT conversations/{id}/ tag method has been changed into PUT conversations/{externalid}/ tag
- PUT conversations/{id}/ remove-tag method has been changed into PUT conversations/{externalid}/ remove-tag
- PUT conversations/{id}/ attached-data method has been changed into PUT conversations/{externalid}/ attached-data
Knovvu Analytics – Fennec
- POST Fennec/device API requires form data instead of raw body.
Enhancements
- Sentiment analysis of customer and the agents are issued by the sentiment service. ML Sentiment service should be deployed, and its URL should be addressed to the inventories. This feature is used for chat and voice call type of conversations.
- Users can search with external id with conversations quick search field.
- Automatic Evaulation scores are now visible in conversations page.
- New Built-in scenarios in Arabic.
- New Acoustical parameter in queries: Agent hesitation count per minute.
- Filtering according to the tags is enabled in statistical comparison and analysis.
- Hold durations are not counted as silence now.
- Video call record size has been limited with 1GB
- Special characters can be added into reasons now.
- Assignments can be created by considering user groups instead of the departments now.
- 3, 6 and 12 months auto deletion periods options has been added to the settings.
- Chat transcripts are enabled to be editable via UI.
- Evaluation results can be filtered by the passive forms as well.
- Interruption events are shown on the conversation waveform now.
- Filtering by scenarios in Real-time notifications is enabled.
- Conversation timeline for chats is presented in conversation details.
- Net Promoter Score (NPS) graph has been added to the dashboard.
- Time periods of trend graphs can be changed for daily, monthly and yearly.
- New built-in scenarios: detecting abnormal monotonicity, speed and negative customer sentiment.
- Non-interactive agent: Agents who does not required to be enter Knovvu Analytics but their conversations are the subject of the analysis can be created by without defining an email or password. The one should add a new claim with type: role and Value: Non-interactive to the client to use relevant API method.
- Reindexing: To restart indexing process of the queries an action button has been added.
Improvements
- Queue broker has been implemented in Analysis Worker Service
- GET Video-Calls list method has been added to API.
- New conversation source: Knovvu Virtual Agent.
- Real-time notifications appear more in UI.
- New permission: Can evaluate their own conversations
- Transcripts can be received from API via GET /conversations/{externalId}
- The appearance of the duration has been revised in UI
- Calculation of NPS has been revised.
- The appearance of the conversation details in the evaluation results has been revised.
- Menu icons has been improved
- Minor UI improvements in statistical comparison and statistical analysis pages.
- Case sensitivity in the queries has been resolved.
- Minor grid optimizations for profile pictures.
- Background worker’s events are kept with a trace id and State Machine Flow screen in history tab for troubleshooting purposes.
- API Definitions has been improved.
- AI Topic services has been enabled to use considering ingress.
- Minor UI improvements.
- Bug fixes.
Known Issues
- Bug: externalid for Posting chat is nullable.
- Bug: chat content for posting chat is nullable
- Bug: Login with tenant action is not working.
- Bug: Lacking Evaluation form validations.
- Bug: externalid of videocall is now automatically generated.
- Bug: Lacking file type validation for screen records
Was this article helpful?