Speech enhancement in entertainment audio
AssigneeDolby Laboratories Licensing Corporation
InventorHannes Muesch
Life of this patent
Estimating the term…
Abstract
The invention relates to audio signal processing. More specifically, the invention relates to enhancing entertainment audio, such as television audio, to improve the clarity and intelligibility of speech, such as dialog and narrative audio. The invention relates to methods, apparatus for performing such methods, and to software stored on a computer-readable medium for causing a computer to perform such methods.
Claims28
A method for enhancing speech in entertainment audio, comprising
- processing, in response to one or more controls, said entertainment audio to improve the clarity and intelligibility of speech portions of the entertainment audio, said processing including
- varying the level of the entertainment audio in each of multiple frequency bands in accordance with a gain characteristic that relates band signal level to gain, and
- generating a control for varying said gain characteristic in each frequency band, said generating including
- characterizing time segments of said entertainment audio as (a) speech or non-speech or (b) as likely to be speech or non-speech, wherein said characterizing operates on a single broad frequency band,
- obtaining, in each of said multiple frequency bands, a measure of fluctuations in speech levels,
- tracking, in each of said multiple frequency bands, the minimum of the audio level in the band, the response time of the tracking being responsive to said measure of fluctuations in speech levels,
- transforming the tracked minima in each band into a corresponding adaptive threshold level, and
- biasing said each corresponding adaptive threshold level with the result of said characterizing to produce said control for each band.
A method for enhancing speech in entertainment audio, comprising processing, in response to one or more controls, said entertainment audio to improve the clarity and intelligibility of speech portions of the entertainment audio, said processing including
- varying the level of the entertainment audio in each of multiple frequency bands in accordance with a gain characteristic that relates band signal level to gain, and
- generating a control for varying said gain characteristic in each frequency band, said generating including
- receiving characterizations of time segments of said entertainment audio as (a) speech or non-speech or (b) as likely to be speech or non-speech, wherein said characterizations relate to a single broad frequency band,
- obtaining, in each of said multiple frequency bands, a measure of fluctuations in speech levels,
- tracking, in each of said multiple frequency bands, the minimum of the audio level in the band, the response time of the tracking being responsive to said measure of fluctuations in speech levels,
- transforming the tracked minima in each band into a corresponding adaptive threshold level, and
- biasing said each corresponding adaptive threshold level with the result of said characterizing to produce said control for each band.
Dependencies are inferred from the claim text.