×

Audio signal transforming by utilizing a computational cost function

  • US 8,706,496 B2
  • Filed: 09/13/2007
  • Issued: 04/22/2014
  • Est. Priority Date: 09/13/2007
  • Status: Active Grant
First Claim
Patent Images

1. A method of audio signal transforming, the method comprising the steps of:

  • receiving a sequence of time domain multiple digital audio samples representing sound;

    computer-implemented processing the time domain multiple digital audio samples to derive a corresponding sequence of multiple audio pulses in the time domain, each said audio pulse being associated with a characteristic frequency of a single pulse corresponding to a pitch associated with the audio pulse, each said audio pulse having a length corresponding to one pitch period of the respective audio pulse;

    generating for each pulse a sequence of repetitions of said audio pulse, said audio pulse being repeated according to its own characteristic frequency;

    deriving frequency domain information associated with at least some of the sequences of repetitions of said audio pulses, each said sequences of repetitions of said audio pulse being represented as a vector of sinusoids based on the derived frequency, said vector of sinusoids corresponds to a sinusoidal series expansion of the specific audio pulse;

    computer-implemented transforming the sound represented by the time domain digital audio samples by processing the audio pulses using the frequency domain information;

    overlapping the audio pulses in a pitch-synchronous manner;

    transforming each input pulse according to transformation parameters including timbre modifications and estimating a timbre envelope defined by harmonics;

    transforming the timbre envelope using a mapping function; and

    obtaining the transformed timbre envelope at each frequency by computing values of the input timbre envelope at the mapped frequency;

    wherein in the speech transformation an overlapping window is used, a computational cost function which depends on a product of the number of pitch periods and the inverse of the minimum fundamental frequency within the window is determined.

View all claims
  • 1 Assignment
Timeline View
Assignment View
    ×
    ×