Hi there, You could use Mnova which will automatically detect the solvent signals just by running an automatic Peak Picking. Then, in the integral options, you could exclude the solvent signals. Finally, integrate the region of interest and the solvent peaks will not be taken into account for the integration.
You could also use the 'multiplet analysis' tool, which by default does not take into account solvent signals. Here you can see an example of a triplet (which integrates by 1H) which is overlapped with the solvent signal (highlighted in red):
https://db.tt/KbOyaBUb
For further information, please get in touch with Mestrelab support department, here: http://mestrelab.com/support/
Regards, Pablo