EmotionLib v2.5 - Happy New Year
EmotionLib v2.5 Release Notes 🎉
Happy New Year! 🎆
We’re excited to announce EmotionLib v2.5, packed with powerful updates and new features to make media content analysis even more efficient and accurate. This release introduces key advancements and improvements across the library.
What's New in v2.5
-
🆕 SAMP Model Integration:
- The new SAMP model delivers accurate MPAA classifications and ensures effective unsafe content blocking.
-
⚖️ OSentiment as Default:
- OSentiment now replaces the previous Positiveness model by default, offering better sentiment variance and improved precision in emotional tone analysis.
-
🚀 Performance Boost:
- OSentiment is much faster than the past version of Positiveness.
-
⭐ Inference Example:
- Example of a video file inference for quick review.