Signals sinchronizer and its role in automated perceptural voice quality testing
Signals synchronizer matches initial and processed signals in time domain. Input of the synchronizer receives signal segments (pDATA), which duration is equal to VAD (Voice Activity Detection) frame. Criterions of VAD activity for them are specified in the pDATA segments.
Any sound signal can be separated into active and inactive phases. The first corresponds to active sound processes, the latter – to low-level background noise. The elementary way of dividing these two phases is to divide them according to signal energy level. However such approach is not accurate enough. In our approach (http://wordpress.sevana.fi/white-paperautomated-sound-signals-quality-estimation/) we use VAD algorithm presented in recommendation G.723 for this purpose (as a part of VAD vocoder). After filtration the state criterions and signal frames enter the synchronizer, which combines active signal fragments and pauses. The modules use common data: buffer of active etalon signal, buffer of active signal under test, buffer of the etalon signal pause, buffer of signal under test pause and readiness criterion of buffers of active signal and pauses. Besides other blocks there is also a counter of synchronization errors, and the output of the synchronizer corresponds to a pair of buffers with active signals or a pair of buffers with pauses. Both of the blocks of synchronizer can initiate an appearance of a pair of synchronized buffers. The synchronized buffers and the criterion of activity are the input of analytical module.
Misty, isn’t it? Well, when trying to solve a tasks of merging objective and subjective estimation of voice (and not only voice, but also sound!) quality. The main idea about the synchronizer is that it matches two voice or/and audio signals in time domain. These signals are whether the speech model generated for testing purposes and system response, or your etalon signal and system response. Basically, we started talking about one very important system feature – comparison of two audio files quality wise. Next time we’ll discuss the analytical module and its functionality of such a system. Don’t be afraid if this looks too “rocket science” – it will be easier when we get to benefits and applications, but it’s always good to learn the background first. Till next post!
