Steps
Bland - Call Number
Steps
Bland - Call Number
The Bland Call Number step lets you call any phone number with an AI agent that sounds like a human. You can configure your phone agent with different base models, voice models, and behavior.
Bland Call Number Documentation
Options
| Name | Type | Description |
|---|---|---|
| Phone Number | Phone Number | The phone number to call. Country code defaults to +1 (US) if not specified. Formatting is flexible, however for the most predictable results use the E.164 format. |
| Task | Plain Text | Provide instructions, relevant information, and examples of the ideal conversation flow. |
| First Sentence | Plain Text | A phrase that your call will start with instead of a generating one on the fly. This works both with and without wait_for_greeting. Can be more than one sentence, but must be less than 200 characters. |
| Wait For Greeting | Boolean | Should the AI speak first or wait for someone else to talk? |
| Request Data | Object | A JSON object that will be sent to bland when the call is answered. |
| Interruption Threshold | Integer | When you increase the interruption latency, you force the AI phone agent to listen longer before responding. In practice, increasing the threshold results in less interruptions and more latency. |
| Pathway ID | Plain Text | This is the pathway ID for the pathway you have created on our dev portal. You can access the ID of your pathways by clicking the ‘Copy ID’ button of your pathway. |
| Model | Dropdown | Select a model to use for your call. Options: gpt4, base, turbo and enhanced. |
| Record | Boolean | To record your phone call, set record to true. When your call completes, you can access through the recording_url field in the call details or your webhook. |
| Transfer Phone Number | Plain Text | A phone number that the agent can transfer to under specific conditions - such as being asked to speak to a human or supervisor. |
| Language | Dropdown | Select a supported language of your choice. Optimizes every part of our API for that language - transcription, speech, and other inner workings. |
| Max Duration | Integer | Set the longest you want the call to possibly go in minutes. After the max_duration minutes have passed, the call will automatically end. |
| Answered by Enabled | Boolean | Enables machine detection when the call starts to determine whether the call was answered by a person or a voicemail. |
| Temperature | Decimal | A value between 0 and 1 that controls the randomness of the LLM. 0 will cause more deterministic outputs while 1 will cause more random. |
| Start Time | Datetime | The time you want the call to start. If you don’t specify a time (or the time is in the past), the call will send immediately. |
| Voicemail Message | Plain Text | When the AI encounters a voicemail, it will leave this message after the beep and then immediately end the call. |
| Voicemail Action | Dropdown | This is processed separately from the AI’s decision making, and overrides it. |
| AMD | Boolean | AMD (Answering Machine Detection) mode helps your AI navigate phone trees and IVR systems. If you know your call will hit an automated system you should switch it on. |
Outputs
| Name | Type | Description |
|---|---|---|
| Transcript | Plain Text | The full transcript of the call. Use this for post-call automation or processing. |
| Answered By | Plain Text | Whether the call was picked up by a “human”, “voicemail”, or “unknown”. |
Tips
- Try different Model and Voice Model settings - these can make a big difference on the quality of the call.
- For help in writing good Tasks, refer to Bland’s prompting guide.
- You can view your call recordings in your Bland dashboard.
- If you need complex dialog control, or the agent isn’t following the Task well, try using their Conversational Pathways. You can paste the ID into Respell once it’s set up.