https://retellai.com/?utm_source=discord&utm_medium=seo logo
what it's the best way to test a frech agent if i ...
# ❓・ai-evy-support
n
@AI Evy what it's the best way to test a frech agent if i only speack english
a
Thanks for tagging me @nicottone. I'll look through relevant sources and get back to you shortly.
### Answer For testing a French agent when you only speak English, Retell AI provides multiple testing methods that work regardless of the language barrier [(1) ]() : LLM Playground is your best option for text-based testing [(1) ]() [(2) ]() . You can: - Type messages in French (using a translator if needed) and observe the agent's French responses [(2) ]() - Test function calling to verify the agent triggers correct functions with proper parameters - Save test cases for regression testing - Reset conversations and iterate quickly LLM Simulation Testing allows automated testing with predefined French scenarios [(1) ]() . This provides: - Batch testing capabilities - Success metrics and scoring - Consistent quality assurance without needing to speak French yourself Web Call/Phone Call Testing validates the actual voice interactions , letting you: - Test voice quality and latency - Validate interruption handling - Verify real-world performance The recommended workflow is to start with the LLM Playground for development , move to Simulation Testing for QA , then use Web/Phone Call Testing for final validation .
Feel free to tag @AI Evy with additional questions. I also appreciate feedback, it helps me improve.
​ To learn more, see the sources I considered: [[1] Testing Overview](https://docs.retellai.com/test/test-overview) [[2] Manually test your agent](https://docs.retellai.com/test/llm-playground) ​
n
@AI Evy can the information in knowledge base be in english?
a
Thanks for the follow-up, @nicottone. I'll get back to you soon.
### Answer Yes, you can have English content in your Knowledge Base even when building a French agent [(1) ]() [(2) ]() . The Knowledge Base stores your source material (webpages, documents, or custom text) , and the agent will retrieve relevant information from it during conversations . For a multilingual setup like French, a community discussion suggests using a multilingual language model and prompting the agent to speak in the target language [(3) ]() . The prompt approach mentioned was: "ask the user their language preference in english and spanish, and once they have told you, use this consistently throughout your conversation" . You could adapt this for French. Another user confirmed that their Spanish bots are "all written in English with instructions to speak Spanish" and it "works perfect" . This indicates you can maintain English content in your Knowledge Base and prompts while instructing the agent to communicate in French. The Knowledge Base content is automatically retrieved and fed to the LLM as context [(1) ]() , so the LLM will translate and respond appropriately based on your language instructions in the prompt.
​ To learn more, see the sources I considered: [[1] Knowledge Base](https://docs.retellai.com/build/knowledge-base) [[2] Empower AI with Dynamic Knowledge Bases | Retell AI](https://www.retellai.com/features/knowledge-base) [[3] Discord message from @naseem_53863 on 2025-02-27](https://discord.com/channels/1186424376300679209/1186462578310910112/1344774325789790310) ​
Feel free to tag @AI Evy with additional questions. I also appreciate feedback, it helps me improve.