Adobe, Inc.

Martin Schmitz

Sven Duwenhorst

Hartmut Warncke

Sven Winter

A method for analyzing media assets such as video and audio files. The method includes providing access to all the frames of a digital media asset. The method includes, with a microprocessor, running a raw analyzer modules to analyze the asset frames to produce sets of raw analyzer result data that are stored in a data cache in a file associated with the asset. The sets of raw analyzer results are linked to the raw analyzer modules with unique identifiers. The digital media asset is played for the raw analyzer modules, which concurrently analyze the temporally-related frames. The raw analyzer results are stored as data tracks that include metadata for the asset such as immutable parameters including histograms. The method includes using a feature algorithm module to generate an analysis result, such as face identification, for the digital media asset based on the raw analyzer results accessed by the identifiers.

Framework for combining content intelligence modules

One exemplary embodiment involves receiving identifications of audio problems in a segment of audio and identifications of corrections for applying to attempt to correct the audio problems, wherein the audio problems were identified by a device applying one or more audio problem detection algorithms to the segment of audio. The exemplary embodiment further involves displaying a user interface comprising representations of the audio problems and representations of the corrections and, in response to receiving a command through the user interface to initiate application of a correction of the corrections, initiating application of the correction.

Systems and methods for semi-automatic audio problem detection and correction

Systems and methods to automatically equalize coloration in speech recordings is provided. In example embodiments, a reference spectral shape based on a reference signal is determined. An estimated spectral shape for an input signal is derived. Using the estimated spectral shape and the reference spectral shape a comparison is performed to determine gain settings. The gain settings comprise a gain value for each filter of a filter system. Using gain values associated with the gain setting, automatic equalization is performed on the input signal.

Growing community of inventors

Braunschweig, Germany

Martin Schmitz