Nodes are the building blocks of a flow. The Nodes toolbox in the flow designer only shows the nodes that are available for the type of flow:
| Group | Nodes | Types |
|---|---|---|
| Flow Nodes | Start, Queue, Wait for termination, Terminate | all |
| Add Participant, Virtual queue, Callback scheduler | call | |
| Deliver | call, email | |
| Settings Nodes | Set Language, Set Skill, Set VIP Class, Set Signaling Time, Set Customer, Holiday closed, Out of business hours, Manual closed | all |
| Set Voice, Allow recording | call | |
| Extended Nodes | Set Variables, Script code, Branch, Basic Math, Send Email, Start Timer | all |
| Data Access Nodes | Database Read, Attach data | all |
| Media Nodes | Media Playback, Background Music, Text To Speech, Background Text To Speech | call |
| Recording Nodes | Start Recording, Stop Recording, Voicemail | call |
| Recognition Nodes | Speech To Text, DTMF Character, DTMF Text, Word Recognition | call |
| Digit recognition, Digit sequence recognition | chat | |
| Value Recognition | all | |
| Text Based Nodes | Move to Folder, Wait for reply | email, fax |
| Automatic answer | email, fax, chat | |
| Event Nodes | State Change Event, Queue Exit Event, End Event, Timer Elapsed Event | all |
What applies to all nodes
- Display Name: the first property of every node, see Naming nodes.
- Fields: Many fields accept either a fixed value or a variable. Enter time spans as days:hours:minutes:seconds. In fields, the flow only replaces its own variables, see Placeholders.
- Outputs: If the flow takes an output that has no connection, the path ends there. End nodes — Queue, Virtual queue, Terminate and Wait for termination — have no output.
- Answering calls: Nodes that need audio answer a call that has not been answered yet, see When a call is answered.
- States: While a node is running, the conversation has a state of its own that supervisors and the Running conversations list can see — for example Announcement, DTMF input, Recognition, Voicemail, Automatic answer, Wait for reply or Sending email. Afterwards, it returns to its previous state.
Nodes with Azure
Speech synthesis, speech recognition and translation run via Microsoft Azure and require the AI Key:
| Without the AI key | Consequence |
|---|---|
| Text To Speech, Background Text To Speech, media cases made from text | It stays silent; the flow continues. |
| Word Recognition, Value Recognition in a call | The node takes Timeout immediately, without an announcement. |
| Speech To Text | The variable remains empty; the flow continues. |
| Speech recognition with DTMF Character and DTMF Text | Only keys count. |
| Text recognition with Digit recognition and Digit sequence recognition | Only digits count, no number words. |
| value types of Value Recognition that Azure recognizes | They find nothing, see Value Recognition. |
| boilerplates that do not exist in the language of the conversation | Instead of a translation, the first available language is used. |
Flow nodes
Start, Queue, Wait for termination, Terminate, Add Participant, Deliver, Virtual queue and Callback scheduler — with properties, outputs and effect.
Settings nodes
Setting the language, skill, VIP class, signaling time, customer, closed states, voice and consent to recording for a conversation.
Extended nodes
Setting variables, making decisions with script code and branches, calculating, sending emails and starting timers.
Data access nodes
Using Database Read to fetch customer data from an external database and Attach data to attach values to the conversation that the agent sees.
Media and recording nodes
Playing announcements, background music and text read aloud, recording calls and recording voicemails — the nodes for call flows.
Recognition nodes
Recognizing keys, spoken words and values — in calls, chats and emails — and using the result in variables or branches.
Text-based nodes
Answering emails, faxes and chats automatically, moving emails to folders and waiting for the customer's reply.