Back to evaluations
Draft evaluation

Build Multilingual Voice Health Assistant with AutoGen and Cartesia — evaluation

Validates multi-agent conversation correctness and Cartesia audio stream synthesis quality.

Evaluation type
task based
Challenge
Build Multilingual Voice Health Assistant with AutoGen and Cartesia
Difficulty
Advanced
Rigor
Not declared

The author has not specified a rigor level.

Open linked challenge

Evaluation overview

How the linked challenge is judged: tasks, benchmarks, and criteria count.

Tasks
1
Benchmarks
0
Criteria
0

Task templates

Inputs and expected outputs.

Task 1

voice_health_triage

Evaluates medical query interpretation and response voice synthesis parameters.

Input format

JSON string containing citizen query in Hindi/Tamil text

Output format

JSON containing recognized medical concept, remedies, and audio stream URL