Search results are generated by SpeechMiner when you do one of the following:
The Search Results grid contains detailed information about each interaction.
You can modify the display to suit your requirements by selecting the type of interactions to view and the columns you want to include in the grid.
From the following menu above the Search Results list, you can select to view:

In some systems, additional columns can be opened for specific types of metadata. For example, if your system stores a "Location" field for calls, a Location column can be added whenever calls are listed in an Interaction Grid. Because columns are added for all selected fields that are relevant to all the types of interactions included in the grid, some columns may be blank for some interactions. For example, Duration is only defined for calls, and Sender is only defined for e-mails. For other types of interactions, these columns are blank.
For information about how to view and listen to the different types of interactions, refer to the Media Player page.
The following columns can be displayed in the Search Results grid. The availability of each column depends on the type of search results list you selected to view:
| Column | Description | Available in: |
|---|---|---|
| Agent | The agent who handled the interaction. |
|
| Audit Time | The time at which the event was audited. |
|
| Auditor | The name of the auditor. |
|
| Bcc | Individuals to which the email interaction was sent to secretly. |
|
| Cc | Individuals to which the email interaction was sent to as a copy. |
|
| Category | The Category or Categories that were assigned to the interaction. If only one Category was assigned, the name of the Category appears. If multiple Categories were assigned, the number of Categories appears; place your mouse pointer on the value to see a list of the Categories. |
|
| Confidence | A value indicating how reliable the identification of Events in the call is. The higher the number, the greater the degree of certainty. You can use this information to choose calls for auditing. Sort the list by this column (by clicking in the column title) and select those calls whose Confidence value is below a certain threshold. Note: This number is not a percentage, but it does range from 1 to 100. |
|
| Date/Time | The start date and time (in hours:minutes) of the interaction. |
|
| Duration | The duration of the interaction, in hours:minutes:seconds |
|
| End Time | The end time of the Event, offset from the start of the call, in hours:minutes:seconds. |
|
| External ID | A unique number assigned to the interaction by the external recording or storage system. |
|
| ID | A unique number assigned to the interaction in the SpeechMiner database. |
|
| Media | The type of interaction found. For example, call, email, video, etc. |
|
| Open | A Play |
|
| Program | The program that was used by SpeechMiner to analyze the interaction. |
|
| Receiver | The name of the individual(s) to whom the email interaction was sent. |
|
| Segment ID | The segment of the audio interaction associated with the event. Segment ID is only available in a Recording UI environment and a Recording + Analytics environment. |
|
| Sender | The name of the individual who sent the email interaction. |
|
| Size | The size of the video file. |
|
| Start Time | The start time of the screen recording in hours:minutes:seconds. |
|
| Stop Time | The end time of the screen recording in hours:minutes:seconds</tt |
|
| Subject | The subject of the email interaction. |
|
| Term ID | The ID number of the term that was identified in the Topic. |
|
| Text |
Note: This column is only visible in grids that contain search results from searches that included terms.
|
|
| Topic |
|
|
| TP / FP | If the Event has been audited, one of the following values is selected, and the entire cell is color-coded:
If you have Event Auditor permissions, you can select a value in this column. To select a value, click it. To clear the selection, click Clear. Note: If you selected Exclude events that have been audited before you initiated your search, previously audited Events will not appear in the grid. |
|
| Type | The video format. |
|
| Workgroup | The workgroup of the agent who handled the interaction. |
|
The following sections represent four common grid procedures.

If the Interaction Grid was generated in response to a search, the upper-left corner of the Interaction Grid contains a summary of the search criteria and settings that were employed during the search. Hover over the summary to see more details about the search criteria.
In addition, if the grid was generated in response to a search for terms, the system may list a "related search" term below the search criteria. For additional information, see Related Searches.
Furthermore, if the system determines that one or more of the search terms was spelled incorrectly, it will correct the apparent errors and indicate this in a note below the search criteria. For additional information, see Spelling Corrections.
To open an interaction click the play icon next to the interaction you want to playback.
For example, to open a call, click the
.
If the interaction is a call, the Media Player opens above the grid, and the call automatically begins to play.
If the interaction is not an audio call, the transcript of the interaction is opened.
Calls that were imported from the Genesys Interaction Recording system may be divided into segments. When this is the case, a
.
To open a segmented call:
The call row is expanded, and a list of the call's segments is displayed in it. Each segment is identified with a
icon. In addition, the Media Player opens above the grid, and the first segment of the call automatically begins to play. Playback stops when the end of the segment is reached.
An event occurs when SpeechMiner identifies a topic in an interaction. Each topic event identified by SpeechMiner has a start time, end time,a name and a type. For example, if the "can offer you a loan" phrase associated with the Loan Offer topic is identified, the start of the event is "can" and the end of the event is "loan". SpeechMiner can also identify non-linguistic events. Non-linguistic events are sounds other than speech (for example, a dial tone, silence, music or noise, key press, and so on).
Event search results are different from a general list of search results in the following ways:
| When you search for events you must search for a term, term ID or a topic |
Each linguistic event recognized in an audio interaction can be audited. This means that a SpeechMiner user listens to the event and then rates the accuracy of the identification. Auditing is an important part of fine-tuning the SpeechMiner's speech-recognition accuracy.
In order to audit an event, the auditor listens to the event, and then rates the accuracy of the event recognition by selecting one of the following values in the TP/FP column of the Event Grid.:
Summary information about event auditing in your system can be seen in the Audit Analysis and Topic Analysis - Audits reports. Existing audit data can be managed globally in the Calls Admin screen.
The Search Results grid contains a list of all the screen recordings in the system.
Each screen recording is a video of the agent's screen during the agent/customer interaction.
Screen recording search results are different from a general list of search results in the following ways:
To view a specific screen recording in the context of the interaction it was recorded, switch to the Interactions list and search for the interaction ID associated with the specific screen recording. The interaction ID appears in the ID column of the Screen Recordings Search Results grid.
From the Search Results grid you can perform a number of batch actions for selected interactions.
The following sections represent the available batch actions.
You can export and save selected audio interactions by creating a zip file that contains the selected interactions.
The zip file contains the following:
To export an audio interaction(s):
| If you select to export a text interaction(s), a text file is created. The text file includes the content of the selected interaction(s). |
A dialog box opens and asks if you want to open or save the .zip file.
You can send links to interactions to other users in the system. Each link is sent in a separate message that appears in the recipient's My Messages widget in the Dashboard page. When the recipient clicks the message, a SpeechMiner Media Player window opens and plays the interaction.
You can also send the recipient an e-mail notification with links to the selected calls. When the recipient clicks a link, the Media Player opens in a new browser tab or window and plays the interaction. If the recipient is not logged into SpeechMiner, the user is prompted to do so before the Media Player opens.
To forward an audio interaction to another user:
An Interaction List is a list of audio interactions that is saved for later use.
SpeechMiner supports two types of Interaction Lists:
When working with interaction lists:
The selected interactions are added to the selected list.
For more information about creating a new list or adding to an existing list, refer to: Interaction Lists.The selected interactions are added to the selected coaching session.
For more information about creating a new coaching session or adding to an existing coaching session, refer to: coaching.Reprocessing enables you to reapply the SpeechMiner recognition process on selected interactions. This procedure is helpful when you know that topic definitions have changed.
To reprocess an interaction(s):
| Reprocessing interactions is only available in an Analytics Only environment. |
You can delete any interaction in the Search Results grid if you have the permissions to do so.
To delete an interaction(s):