### Answer I don't have specific documentation about the exact "your voice cut out" phrasing you saw in Retell's demo. However, based on the available sources, here's how you can address poor audio quality: Prompt Engineering Approach The behavior you're describing is typically implemented through prompt engineering [(1) ]() . You can add instructions to your agent's prompt that tell it to handle unclear responses by asking for clarification. For example, you could include guidance like: "When you cannot clearly understand the user's response, especially for critical information like date of birth, politely acknowledge the audio issue and ask them to repeat. Use phrases like 'I'm sorry, I didn't quite catch that' or 'Your audio cut out for a moment—could you repeat your date of birth?'" Fallback Strategies Retell AI supports fallback strategies that incorporate prompt engineering techniques to clarify user intent when uncertainty arises . When the AI is unsure of a user's meaning, it should ask follow-up questions to double-check and prevent misunderstandings, ensuring a more accurate and helpful response . Technical Factors Affecting Recognition Keep in mind that Retell AI's ASR (Automatic Speech Recognition) is designed to be noise robust, with the ability to filter out background noise, accents, or speech variability for clean transcriptions [(2) ]() . However, extremely poor audio quality can still challenge even robust ASR systems . For specific guidance on implementing this exact behavior in your agent's configuration, I recommend reaching out to Retell support with your use case.

​ To learn more, see the sources I considered: [[1] Troubleshooting AI Voice Agent Issues | Retell AI](https://www.retellai.com/blog/troubleshooting-common-issues-in-voice-agent-development) [[2] Automatic Speech Recognition (ASR)](https://www.retellai.com/glossary/automatic-speech-recognition-asr) ​

Feel free to tag @AI Evy with additional questions. I also appreciate feedback, it helps me improve.