Pitch-synchronous speech analysis

In speech synthesis, it is common to perform pitch-synchronous analysis, where the frame... 

We have just seen that elosed phase linear prediction requires that we analyse eaeh pitch period separately. This type of speech analysis is called pitch synchronous analysis and can only be performed if we are in faet able to find and isolate individual periods of speech. We do this by means of an pitch marking orepoch detection algorithm (EDA). 

The above technique bears some similarities to the TD-PSOLA technique in that it uses a pitch-synchronous analysis to isolate individual pitch periods, after which modification and resynihesis is performed. In fact in a technique called linear prediction pitch synchronous overlap and add or LP-PSOLA, we can use the PSOLA more or less directly on the residual rather than the waveform. As above, epoch detection is used to find the epochs. The residual is then separated into a number of symmetrical frames centred on the epoch. Pitch modification is performed by moving the residual frames closer or further away, and duration modification is performed by duplication or elimination of frames, in just the same way as in TD-PSOLA. The only difference is that these operations are performed on the residual which is then fed into the LP filter to produce speech. This technique differs only from the Hunt technique in the shape of the frames. Both techniques uses window functions with their highest point at the epoch in Hunt s technique the windows are asymmetrical with the idea that they are capturing a single impulse, in LP-PSOLA the windows are symmetrical. In listening tests, the two techniques produced virtually identical quality speech. 

In speech synthesis, it is common to perform pitch-synchronous analysis, in which the frame is centred around the pitch period. Since pitch is generally changing, this makes the frame shift variable. Pitch-synchronous analysis has the advantage that each frame represents the output of the same process, that is, the excitation of the vocal tract with a glottal pulse. In unvoiced sections, the frame rate is calculated at even intervals. Of course, for pitch-synchronous analysis to work, we must know where the pitch periods actually are this is not a trivial problem and wftl be addressed further in Section 12.7.2. Note that fixed-frame-shift analysis is sometimes referred to as pitch-asynchronous analysis. 

Pitch-scale modification is performed 1 recombining the frames on epochs that are set at different distances apart from the analysis epochs, as shown in Figure 14.3. All other things being equal, if we take for example a section of speech with an average pitch of 100 Hz, the epochs will lie 10 ms apart. From these epochs we perform the analysis and separate the speech into the pitch-synchronous frames. We can now create a new set of epochs that are closer together, say 9 ms apart. If we now recombine the frames by the overlap-add method, we find that we have created a signal that now has a pitch of 1.0/0.009 = 111 Hz. Conversely, if we create a synthetic set of epochs that... 

PSOLA, which operates in the time domain. It separates the original speech into frames pitch-synchronously and performs modification ly overlapping and adding these frames onto a new set of epochs, created to match the synthesis specification. Residual-excited linear prediction performs LP analysis, but uses the whole residual in resynthesis rather than an impulse. The residual is modified in a manner very similar to that of PSOLA. 

---