Delivering fundamental frequency and amplitude envelope cues to enhance speech understanding
AssigneeArizona Board of Regents for and on behalf of Arizona State University
InventorsSid P. Bacon, Christopher A. Brown, Frederic Apoux
Life of this patent
Estimating the term…
Abstract
Systems, apparatus, and techniques are described to perform operations including receiving speech that includes audio signals from a talker. A fundamental frequency, unique to the talker, is extracted from the audio signals. A tone in frequency with the extracted fundamental frequency of the received speech is modulated and the modulated tone is delivered to an audio transducing device.
Claims13
A computer-implemented method to enhance speech understanding, the method comprising:
- receiving speech from a talker at a system including one or more computers, the speech including audio signals;
- extracting, at the system, a fundamental frequency from the audio signals, wherein the extracted fundamental frequency is unique to the talker;
- modulating, at the system, a tone of the speech in frequency with dynamic changes of the extracted fundamental frequency of the received speech, wherein the dynamic changes are obtained using YIN algorithm;
- and
- delivering the modulated tone as an acoustic signal.
A computer-readable medium tangibly encoding software instructions executable by one or more data processing apparatus to perform operations comprising:
- receiving speech from a talker, the speech including audio signals;
- extracting a fundamental frequency from the audio signals, wherein the extracted fundamental frequency is unique to the talker;
- modulating a tone of the speech in frequency with dynamic changes of the extracted fundamental frequency of the received speech, wherein the dynamic changes are obtained using YIN algorithm;
- amplifying the modulated tone;
- and
- delivering the modulated tone as an acoustic signal.
A cochlear implant system comprising:
- a cochlear implant;
- means for receiving speech from a talker, the speech including audio signals;
- means for extracting a fundamental frequency from the audio signals, wherein the extracted fundamental frequency is unique to the talker;
- means for modulating a tone of the speech in frequency with dynamic changes of the extracted fundamental frequency of the received speech, wherein the dynamic changes are obtained using YIN algorithm;
- and
- means for providing the modulated tone to the cochlear implant as an acoustic signal.
A system comprising:
- a first transformer to receive an audio signal and transform the received audio signal into a frequency-domain representation;
- a processor operatively coupled to the transformer, the processor including:
- a frequency compression unit to perform frequency compression of the frequency-domain representation of the audio signal,
- a frequency transposition unit operatively coupled to the frequency compression unit to perform frequency transposition of the frequency-domain representation of the audio signal on which the frequency compression has been performed;
- a second transformer operatively coupled to the processor to transform the compressed and transposed audio signals into a time-domain representation;
- a transmitter operatively coupled to the second transformer to deliver the signals in the time-domain representation as acoustic signals;
- an amplifier operatively coupled to the second transformer to amplify the signals in the time-domain representation;
- and
- a level control unit operatively coupled to the amplifier to control a level by which the amplifier amplifies the signals in the time-domain representation.
Dependencies are inferred from the claim text.