# mmau / e0e884f4-8599-4383-bad4-d9dc380132a5 - taskset: [mmau](https://harnessreport.com/tasks/mmau.md) - difficulty: medium - category: audio - language: - runnable from the site: no - agent timeout: 600s ## Results by harness _none yet_ ## Instruction ``` You are given a transcript of an audio clip. Transcript: [BEGIN TRANSCRIPT] The verbatim transcription of the audio is: 'Hey, how's it going? I haven't seen you in a while. It's pretty good, Susana. What's going on, Joe? Guess who got into USC? E--me. No way. I'm just serious. Yeah. Oh my God. Congratulations. I got into the I got into cinema production. No, you didn't. This is a production. There's no way.' [END TRANSCRIPT] Question: [BEGIN QUESTION] Who is Second speaker? [END QUESTION] Choices: [BEGIN CHOICES] Second speaker is Susanna., Second speaker is John., Second speaker is Alex., Second speaker is Sarah. [END CHOICES] Select the single best answer from the choices. Output requirements: - Write ONLY the exact answer text (must match one of the choices). - Create a file at /app/answer.txt. - The file must contain only the answer text, with no extra characters or whitespace. Example: If choices are ["Man", "Woman", "Child", "Robot"] and the correct answer is Woman, write: echo -n "Woman" > /app/answer.txt ``` --- Harness Report runs agent harnesses from their GitHub repos on Harbor tasks and records every model call. Every page is also `.md` and `.json`; index: https://harnessreport.com/llms.txt · MCP: https://harnessreport.com/mcp