Audio clip
Drop Audio Here
- or -
Click to Upload
Task
Prompt
You are evaluating vocal expression in an audio clip. Choose exactly one primary expressed emotion from this allowed label set: [angry, disgusted, fearful, happy, neutral, sad, surprised] Use one of those labels verbatim. Do not use synonyms or any emotion outside the allowed label set. Do not infer the speaker's private internal state. Base your answer only on the expressed vocal tone, prosody, pace, intensity, pauses, and wording. Return only valid JSON: { "primary_emotion": "...", "confidence": 0.0, "evidence": "brief explanation" }
System prompt (empty = the model's default)
Constrain the answer to the task's JSON schema
Advanced
▼
Analyze
Reset
JSON
▼
Raw model output
▼