Prosodyanalyzer
Single-turn prosody and emotion analysis over speech WAV clips. Model returns JSON with four acoustic metrics and a style label.
- Domain
- rl-env
- License
- unknown
- Published
- Oct 2025
Cite
Notes
Only stored in your browser.
Top score 51.9% by Gemini 1.5 Flash (Sep '24) - 5 models reporting
Score history
2Top models
5
Related tools
1Implementations, trainers, datasets and scaffolds linked to this eval.
FAQ
- What is Prosodyanalyzer?
- Single-turn prosody and emotion analysis over speech WAV clips. Model returns JSON with four acoustic metrics and a style label.
- What is the current top score on Prosodyanalyzer?
- The top reported score is 51.9% by Gemini 1.5 Flash (Sep '24), across 5 models reporting.
- How can a model improve its Prosodyanalyzer score?
- Tools linked to Prosodyanalyzer on Sophon include Prosodyanalyzer RL Env (Community) - RL environments, datasets, and scaffolds that target this eval.
- What license is Prosodyanalyzer under?
- Prosodyanalyzer is available under unknown.