Imagicle Voice Analytics Admin Guide
The access to Imagicle Voice Analytics is available within Imagicle UCX Platform web portal, by selecting relevant option. By leveraging their standard login credentials, with or without SSO, each user accesses the UCX Platform web portal with own, specific permission level.
Username is always the email address. If your company is leveraging Single Sign-On authentication, then your login is redirected to your Identity Provider for multi-factor authentication, where applies.
IMPORTANT Based on your job function, you can have access to Voice Analytics interface as Administrator, Controller or standard User. Controller can view and transcribe recordings performed by the whole company, Administrator can provision new users and view/transcribe all recordings, while a standard user can view/transcribe own recordings only.
Voice Analytics Dashboard
Once authenticated, user accesses to Voice Analytics Dashboard page, which shows statistical data about the users’ behaviors, in terms of sentiment analysis trends. Below you can see an Imagicle Voice Analytics Dashboard sample:
On top page header, you can apply multiple filters on collected conversations and relevant sentiment data:
Date filter → Filter by a predefined time interval or by a custom date range. See below:
Users’ filter → Allows to apply a filter to list to recorded conversations uploaded by specific users only. See below:
Funnel icon → to remove all configured filters.
Analyzed data involves all selected users and it provides three “widgets” including different data:
Overall Sentiment
Embedded gauge graph represents sentiment average level of those conversations included in applied time frame, with an incremental/decremental percentage against previous adjacent time frame.
In the lower part of the widget, three counters are respectively showing the number of negative, neutral and positive calls.
If you hoover mouse pointer on this widget, additional data is displayed:
Applied time frame and adjacent previous time interval, considered to calculate the incremental/decremental percentage.
An up or down arrow indicating an overall increase (more positive) or decrease (more negative) of sentiment value.
Sentiment value range, between -1 and +1
Sentiment Breakdown
Embedded slider graph represents sentiment average levels of those conversations included in applied time frame, referred to local party (user) and remote party.
If you hoover mouse pointer on this widget, additional data is displayed:
Applied time frame.
Sentiment value range, between -1 and +1
Sentiment Trend
This widget includes a graph representing sentiment trends of those conversations included in applied time frame, at overall, local party (user) and remote party selectable levels.
X-axis represents selected time frame, where a proper time unit is applied depending on filtered time range: hour, day, week, month.
On top of the widget you can find three buttons, to toggle Overall, Users and Remote parties sentiment graph patterns. By default, Overall pattern is displayed, but you can turn on all of them or just select the required one(s).
COMPARE switch on top-right allows to show or hide trend patterns related to previous adjacent time frame. Previous time frame trend patterns are displayed with dotted lines. If you hoover mouse pointer on this widget, additional data is displayed:
Applied time frame and adjacent previous time interval, considered to display relevant trend patterns.
Placing mouse pointer on X-axis time units intersecting trend patterns, pops-up a detail panel including filtered time frame and sentiment values of selected graph patterns (Overall, Users and Remote parties) in the specific date/time.
Other menu options
On left-side bar, four options are available:
→ to access Sentiment dashboard (default landing page)
→ to access recorded conversations list
→ Manage Alarms
→ User Groups management
On top-right corner, you can find the logout icon
Voice Analytics Recordings Page
Next available menu item is Voice Analytics Recordings page, which shows the list of all available recordings performed by one or multiple users. Below you can see an Imagicle Voice Analytics Conversations page sample:
For each recording entry, the following data is available:
Dir → Call direction: = Inbound; = Outbound
From → Calling Party name or phone number
To → Called Party name or phone number
Time → Date/time stamp
Info → Icon to indicate transcription status; emoticon to indicate evaluated overall sentiment analysis status; icon indicates a screen recording is available for this call. Icon indicates that a text note has been added to this recording; If these icons appear, then the call is too short for any evaluation analysis.
Duration → Call duration
Tags → Matched tag(s) inside the conversation
Delete → Trash icon to delete a recording entry
On top page header, you can apply filters on conversations list, similar to those highlighted in Dashboard section:
Keyword/sentence search à allows to search for specific text across all transcribed recordings
Sentiment filters (multiple options can be selected):
→ Any sentiment
→ Negative sentiment calls only
→ Neutral sentiment calls only
→ Positive sentiment calls only
→ Sentiment data not available
MORE FILTERS → allows to apply additional search criteria across all imported/transcribed recordings. See below:
Recorded MP3 audio file upload
Imagicle UCX Suite ver. 2022.Winter.1 and above allows to automatically upload all recorded conversations to Voice Analytics web portal. Nevertheless, if you are leveraging an older UCX Suite version or you have some recordings performed before purchasing Voice Analytics subscription, you still have the chance to manually upload these MP3 audio files, once exported from Imagicle Call Recording “Search & Play” web portal or gadget. Please consult Imagicle UCX Suite’s Call Recording User Guide for more details.
The button is available under the recording filters section. Once hit, below dialog pops-up:
FILE UPLOAD → Drag & drop your MP3 file in this area or click to open file system dialog to locate the file within your workstation.
SELECT LANGUAGE → Conversation language to be transcribed. You can manually choose language or let internal engine detect the correct one.
MAX NUMBER OF SPEAKERS → Number of call parties. You can manually select the number of parties or let internal engine detect the correct one.
When a new MP3 file is uploaded, Voice Analytics engine immediately starts the transcription and sentiment analysis, which might require few minutes, depending on call duration.
Please note that foreign MP3 files, not coming from Imagicle Call Recording application, or files greater than 25 Mbytes are rejected by Voice Analytics engine.
Please remember that Voice Analytics subscription is based on overall amount of analyzed conversation minutes available to each user or company-wide. If the maximum amount of minutes has been reached, the application might refuse new recordings automatic/manual upload. This information is available on top header of each Voice Analytics web page. See below sample:
Voice Analytics Recording Transcription and Sentiment Analysis
From Voice Analytics Recordings page, you can click on any available recording to access relevant audio transcription and sentiment analysis. See below screenshot sample:
On top page header, you can find:
Keyword/sentence search → allows to search for specific text within selected recording. See below:
Panel including date/time stamp, call parties names/phone numbers, overall sentiment and a button to export the transcription content. By clicking icon, the following pop-up appears, allowing to download the transcription as text file:
Under recording summary panel, you can find a graphic pattern representing audio recording over a blue background. This is actually an audio player, which allows to listen to the whole recording content. If a screen recording is associated to the call, a “ VIEW SCREEN RECORDING” link appears underneath the audio player. If you click on it, a new video viewer window pops-up, allowing to simultaneously watch audio+screen recording.
Note panel to the right allows the user to add a text note indicating the subject of the conversation and other valuable information. This note is searchable from Recordings page.
Tags panel to the right allows to add specific keywords/sentences tags to the recording, for an easier retrieval of several recordings including one or multiple tags.
Transcript Panel
Transcript Header → Includes call parties names/phone numbers.
Conversion segments → Each leg includes involved call party, a time stamp, play audio button, voice transcription and sentiment analysis emoticon:
Smile (positive) mood
Neutral mood
Sad (negative) mood
The AI engine embedded in Voice Analytics application can analyze the intonation, cadence and transcription content of the whole conversation or single sections, to detect the “mood” of involved call parties and rate it as negative, neutral or positive.
Voice Analytics Alarms Management
From any Voice Analytics web page, you can click on Alarms icon to open relevant web page. See below screenshot sample:
From this page, an Administrator or a Controller can create, review, or delete own alarms.
Create a new alarm
By clicking on green “CREATE ALARM” button or by editing an existing one, the following pop-up windows appears. See below samples:
Please enter an Alarm Name (mandatory), toggle “NOTIFICATION” switch to enable email notifications and select one or multiple sentiment moods where you need to get an alert, or just leave the default “All sentiments” option ( icon). Then you can add one or more words or sentences to be notified if they are included into transcribed recordings. Please hit <ENTER> or “+” to add each keyword or sentence into grey area or click “CLEAR ALL” to remove all keywords. Eventually hit “CREATE” or “SAVE” to save the alert.
You can also choose how keywords/sentences are matched within transcribed conversations, by selecting desired matching criteria from relevant pull-down menu:
Minimum keyword length is two characters and it can’t be duplicated. Search is case insensitive.
It is not possible to add space(s) at the end of a keyword or sentence.
If you are transcribing English or German conversations, you can enter numeric values into keywords/sentences and they are correctly matched. This is not true for other languages, like French where “2” is different than “deux”.
Alarm notification
If there’s a match at sentiment level or keywords/sentences level, then an email notification is sent to the Administrator or Controller who entered the alarm. Two typical email notification samples are shown below:
If multiple alarm rules are matched (also within same conversation), you respectively receive multiple email notifications.
You can click on “ACCESS RECORDING” button to open Voice Analytics web portal to the Recording Transcription page, related to involved recording.
Voice Analytics Groups Management
From any Voice Analytics web page, you can click on Groups Management icon to access relevant web page. See below screenshot sample:
From this page, an Administrator or a Controller can create, review, or delete users’ Groups.
Create a new Group
By clicking on green “CREATE GROUP” button or by editing an existing one, the following pop-up windows appears. See below samples:
Please enter a Group Name (mandatory), and select the supervisors for this group, out of existing VA users. Then you can add one or more words group members, again selected from existing VA users. Eventually hit “CREATE” or “SAVE” to save the Group.