IN speech signal processing, i am getting these two terms more and more. what are they actually?
There are two types of features of a speech signal:
- The temporal features (time domain features), which are simple to extract and have easy physical interpretation, like: the energy of signal, zero crossing rate, maximum amplitude, minimum energy, etc.
- The spectral features (frequency based features), which are obtained by converting the time based signal into the frequency domain using the Fourier Transform, like: fundamental frequency, frequency components, spectral centroid, spectral flux, spectral density, spectral roll-off, etc. These features can be used to identify the notes, pitch, rhythm, and melody.
The most common spectral feature used by beginners in the study of sound is the direction of frequency modulation, rising, level or falling. This helps in the study of tone and intonation. In music or speech acoustics, a Fourier transform produces the harmonic components of the sound called the spectral structure of the signal. This can be displayed in various ways as Caka pointed out above. Thus, in a periodic signal where the vibrations repeat over time, if the first harmonic partial is 100 Hz, for example, the following ones will be 200, , 300, 400, 500, etc. But this is rarely the case in naturally produced sounds. Please note that there is no established law linking spectral components or features to the reality that we here when we experience sound.
I am sharing an example papers which proposes algorithm to extract temporal features.
http://scitation.aip.org/content/asa/journal/jasa/135/1/10.1121/1.4836055
The most successful spectral features used in speech are (i) Mel frequency cepstral coefficients (MFCC) and (ii) Perceptive Linear Prediction (PLP) features. It is well known that the basilar membrane in the inner ear actually analyzes the frequency content of the speech we hear. In fact, the analysis of basilar membrane can be modeled by a bank of constant Q, band pass filters. There also exist the critical bands, which give rise to the phenomenon of masking - where one strong tone or burst can mask another weaker tone within the critical band. Actually, both MFCC and PLP capture these characteristics of our auditory system in some way; so, even though it looks strange, the same features give reasonably good performance for speech recognition, speaker recognition, language identification and even accent identification ! However, these spectral features are not very robust to noise.
On the other hand, some of the time domain (temporal) features such as plosion index and maximum correlation coefficient are relatively more robust to noise.
In more complex signals such as speech or music where the signal properties change over time, it is clearly more meaningful to refer to the changing frequency content over a smaller time interval rather than over an infinite time interval and The short-time Fourier transform plays a significant role in discrete-time processing of speech. By Fourier transform, the spectral features are derived by integrating the temporal envelopes in short-term frames and the temporal features are formed by converting these envelopes into modulation frequency components.
the spectral feature is obtained by converting the time-based signal into the frequency domain using Fourier Transform.
The temporal features are calculated directly on the temporal waveform.
- Badges
- Science topic
- Similar topics
- Correction