Quick overview
This workflow manually triggers two sequential IVR quality-sample calls via the CALL-E API, validating transcripts and metadata post-call. Note: Default placeholders must be replaced with authorized E.164 test numbers before execution.
How it works
- Runs when you start the workflow manually.
- Loads the CALL-E API configuration (base URL, polling interval, timeout) and stops with an error if the API key is missing.
- Creates a small list of IVR QA call tasks (phone, task instructions, locale/region, and metadata) to be processed.
- Processes the call list one item at a time, building the CALL-E
/v1/calls request with a JSON result schema, metadata, and an idempotency key.
- Calls the CALL-E API to create the call and repeatedly polls the call status until it reaches a terminal state (completed/failed/canceled) or times out.
- Extracts the call outcome, structured result, transcript turns, and failure signals, then formats a demo-friendly per-call view and an overall execution summary.
Setup
- Add your CALL-E API key by replacing
replace_with_calle_api_key in the CALL-E configuration step.
- Replace the sample phone numbers with numbers you own or are authorized to call, and adjust the task text and metadata for your IVR QA scenario.
- Review and adjust the CALL-E base URL, polling interval, and timeout values if your environment requires different settings.