Our New Sam Audio Model Transforms Audio Editing
3 days ago
- #creative tools
- #AI audio
- #sound separation
- SAM Audio is a new AI model for sound segmentation, allowing users to isolate specific sounds from audio mixtures using text, visual, or time span prompts.
- It enables professional-grade audio separation for various applications like music, podcasting, film, and scientific research.
- The model supports three prompting methods: text (e.g., 'dog barking'), visual (clicking on objects in videos), and span (marking time segments).
- SAM Audio is the first unified model to match natural human audio processing, offering cutting-edge performance in real-world scenarios.
- Users can try SAM Audio in the Segment Anything Playground, upload their own audio/video, or download the model.