is that possible to find SNR of an environment ? ie., i have a outdoor environment where i cannot find the clean speech ,if so i want to measure the signal power to noise power how to do that

can i do some thing like put a voice activity detection and find voice frames and find the voice frames power to non voice frames power will that be SNR of that environment ?

i have seen some tools like baudline giving segmental SNR will they follow the same method i have mentioned or how this actually happen i have also listened to tools like SOX


1 Answer 1


I think it all depends on the SNR region that you are dealing with. Using a voice activity detector (VAD) seems to be worth an attempt, but consider that VADs fail if the SNR is too low. The most simple approach might be using quantiles of the power distribution. Some lower quantiles will probably not contain power that is caused by the speech signal part while high quantiles will contain power resulting from the speech signal. Choosing the right quantiles might be a challenge, but due to the straight forward approach it might by worth a try...

Furthermore, by reinterpreting your problem to noise floor estimation, you might succeed by using some noise floor estimation algorithm (e.g. Minimum-Statistics by Rainer Martin [1, 2]) and calculating the SNR from that.

[1] Rainer Martin, “Spectral Subtraction Based on Minimum Statistics,” presented at the EUSIPCO, 1994. [2] Rainer Martin, “Noise Power Spectral Density Estimation Based on Optimal Smoothing and Minimum Statistics,” IEEE Transactions on Speech and Audio Processing, vol. 9, no. 5, pp. 504–512, Jul. 2001.

  • $\begingroup$ no problem i have some efficient VAD which can work up to -5dB SNR assuming Efeective VAD presence can you suggest any approach $\endgroup$
    – kakeh
    Oct 18, 2013 at 11:06

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.