The following operators can be used to better focus your queries.
( ) , AND, OR, NOT, W/#
? single char wildcard, not at start
* multi char wildcard, not at start
(Cat? OR feline) AND NOT dog?
Cat? W/5 behavior
(Cat? OR feline) AND traits
Cat AND charact*
This guide provides a more detailed description of the syntax that is supported along with examples.
This search box also supports the look-up of an IP.com Digital Signature (also referred to as Fingerprint); enter the 72-, 48-, or 32-character code to retrieve details of the associated file or submission.
Concept Search - What can I type?
For a concept search, you can enter phrases, sentences, or full paragraphs in English. For example, copy and paste the abstract of a patent application or paragraphs from an article.
Concept search eliminates the need for complex Boolean syntax to inform retrieval. Our Semantic Gist engine uses advanced cognitive semantic analysis to extract the meaning of data. This reduces the chances of missing valuable information, that may result from traditional keyword searching.
A technique is presented for describing the speech signal in terms of a small number of parameters, without excessive loss of information.
English (United States)
This text was extracted from a PDF file.
100% of the total text.
Page 1 of 1
Spectral Parameter Estimator
A technique is presented for describing the speech signal in terms of a small
number of parameters, without excessive loss of information.
Spectral measurements having the dimension of amplitude (e.g., pressure,
voltage, etc.) are easily obtained by known techniques (e.g., filterbank), but
experience has shown that measurements having dimension frequency (e.g.,
formant frequencies) describe the signal more economically, with fewer
A technique for obtaining frequency measurements characterizing a speech
signal is as follows: Step 1. Calculate the logarithmic power spectrum of the
signal (decibels vs. frequency). Step 2. Calculate the Hilbert transform of this
spectrum. This is the phase of the signal if produced by a minimum-phase
network. Step 3. Add a linear function to the phase so as to make the phase at
zero frequency equal to zero, and to give it an average slope of approximately pi
radians per kilohertz. This produces an average spacing of 1 kilohertz between
formants, which is correct for a typical speaker whose vocal tract length is 17 cm.
The slope may be fine tuned for individual speakers. Step 4. Plot the inverse of
this resulting function: frequency vs. phase. Step 5. Sample the function at
convenient intervals, for example, every pi/2 radians.