I am not able to test using Test Suite for both voice and text. In voice transcript, it is confusing between who is user and who is AI and then getting stuck in loop by repeating same question/response. In chat it gets into infinite loop just after any tool call.

Suggestion: To help prevent role confusion and looping in your Test Suite scripts, try explicitly labeling each turn in your script with

[User]

and

[AI]

, and ensure your rubric includes criteria to check for proper turn-taking and tool call completion. For example:

Script:
1. [User] "I need to schedule an appointment."
2. [AI] Confirms appointment time.
3. [User] Accepts or requests alternative.

If the issue persists, consider simplifying your script to isolate the problematic step, and reach out to [Vapi Support](https://docs.vapi.ai/support) with your test case details for further troubleshooting. Sources: - [Test Suites documentation](https://docs.vapi.ai/test/test-suites) - [Chat Testing documentation](https://docs.vapi.ai/test/chat-testing) - [Vapi Support](https://docs.vapi.ai/support)

the same script had worked fine previously

Message marked as helpful by @! AetherMind! 🎉