We've released updates to the /media/enhance v1.1 and /media/analyze b1.4 APIs.
- Music Detection feature is introduced in podcast and voice-over presets. This feature can also be enabled by setting the audio.music.detection parameter to true. When enabled, the Enhance API intelligently adjusts to handle speech and music portions of the content allowing for more optimal results.
- Trim feature allows the Enhance API to be called to return a portion of the enhance content rather than just the whole file.
- Sibilance Reduction has been significantly improved to offer increased reduction of sibilance while preserving speech quality and intelligibility.
- Minimum file duration of 2.0 seconds of content can be analyzed. Previous versions minimum file duration was 1.8 seconds.