We've released updates to the /media/enhance v1.1 and a new version of /media/analyze b1.3.

Changelog

Media Enhance API

• New Speech Plosive reduction detects and reduces excessive plosive sounds within speech -- typically caused when the letter P and B is spoken too close to a microphone.

Media Analyze API

The b1.3 beta release introduces a new method for receiving content summary, event and time series analysis results. With the introduction of this new output format we have also deprecated the previous v1.1 version of the API.

Changelog:

• Analyze requests now require a URL to a writable location where Analyze API will write the results. The results are no longer available in the response of GET command.
• Clipping data reports the locations of clipped sections, enabling you to quickly identify areas within the file that have been recorded to a full scale level (which often indicates distortion).
• Loudness time series data is now added, giving loudness, sample peak and true peak values for every second of the media file. This can be used to understand the locations of loud and quiet sections and the journey of dynamics in a media file.
• Audio sections for content type are now returned, reporting the location of music, speech and silence sections within a file. Each section includes:
  • Start time - the time index (in seconds) when a section starts within the media file
  • Duration - length of a section (in seconds)
• Music Sectional Analysis For each section of music, deeper analysis is now returned to provide directional information. Due to the complex nature of music and its varied expressions, this data should be used only as guidance.
  • Genre - the musical genre of a section of music
  • Era - the decade that a section of music evokes
  • Tempo - the speed or pace (as beats per minute) of a section of music
  • Key - the musical key of a section of music
  • Instrumentation - the top three instruments detected in a section of music