![cepstral voices cepstral voices](https://www.mdpi.com/applsci/applsci-10-00151/article_deploy/html/images/applsci-10-00151-g002.png)
Cepstral analysis can be robust in detecting speech with a missing fundamental frequency. (overtones, by the way, are elements of the signal which are oscillating at multiples of the fundamental frequency i.e. However, for speech, which is a complex signal, there is a large number of overtones.
![cepstral voices cepstral voices](https://imag.malavida.com/mvimgbig/download-fs/cepstral-11427-3.jpg)
If for example, you were to pass a pure sine wave through cepstral analysis, you would get terrible results. Cepstral analysis relies entirely on the plentifulness and strength of the overtones of your signal. What you describe is "cepstral analysis" which is a method mainly used for the extraction of pitch from speech. More on this later, let me answer your other questions. Unless you have some very specific circumstances, I would almost always recommend using autocorrelation. My consensus is that the autocorrelation method is by far the best pitch detector in terms of the tradeoff between accuracy, complexity, noise robustness, and speed. I've done a lot of searching, reading, and experimenting over the past few years. We need tags for vDSP, accelerate framework, cepstral analysisĪlthough I am not an expert and have had minimal formal training, I think I know the best answer to this problem. :| I have suggested to the maintainers that SO keep track of attempted tags, but I'm sure I was ignored. PS I get SO annoyed when I want to create tags, but cannot. The main question: Is it accurate enough? Can the accuracy be improved? I have just been told by an expert that the accuracy IS INDEED not sufficient.
![cepstral voices cepstral voices](https://ars.els-cdn.com/content/image/1-s2.0-S0892199720301028-gr2.jpg)
When it encounters one, it uses a cunning technique (the change in phase since the last FFT) to more accurately place the actual peak within the bin. I have previously used a technique which scours the frequency bins of a single FFT for local maxima. However, I'm wondering about the accuracy of the final result. It looks as though it has functions to handle all of these tasks. I am attempting to achieve this with vDSP library. More to the point, I can't understand why I didn't think of it. I did a lot of hunting and asking questions several weeks worth. I can't understand how I didn't come across this technique earlier. take log of square of absolute value (can be done with lookup table).Someone on IRC just explained to me how taking a double FFT achieves this. I'm looking to extract pitches from a sound signal.