Nothing Special   »   [go: up one dir, main page]

×
Please click here if you are not redirected within a few seconds.
Our feature set consists of two main categories of features. First, we extract two series of prosodic features. Second, we extract two type fractal dimension ...
In this work, we introduce a new speech/non-speech detection system based on fractal dimension and prosodic features plus the common used MFCC features. We ...
People also ask
In this work, we introduce a modified speech/non-speech detection system based on prosodic features characterizing the smoothness of the peak index series ...
This manuscript provides solutions to improve the performance of a speech/non-speech detection system in very noisy environment (for both stationary and short- ...
Non-Speech Segment Rejection Based on Chaotic and Prosodic Features. Article. Full-text available. Soheil Shafiee · Farshad ...
May 3, 2021 · We propose a method which does not require model training and utilizes two prosodic cues that are based on ASR output.
Missing: chaotic | Show results with:chaotic
Speech/non-speech segments detection based on chaotic and prosodic features · Computer Science. Interspeech · 2008.
Abstract. In this work, we describe the use of acoustic-prosodic fea- tures to detect and localize non-native named entities spoken.
Missing: chaotic | Show results with:chaotic
Accurate identification of word fragments will not only improve recogni- tion accuracy, but also be very helpful for dis- fluency detection algorithm because ...
The prosodic features use no word, context, or speaker information. Results with 19% WER speech recognition show improvements from lexical features. (EER=23.1%) ...