Some additional hints:

1. Use also at input of DC/DC-converter such a pi-filter.

2. Some DC/DC-converters show a rather high inrush current at the moment of turn-on. So, some types might refuse to work with a partially discharged battery!
Other types provide a built-in soft-start and might be the better choice. In any case, this is a point to be carefully checked!
Remember: A big cap across battery terminals is always of great help, even if an internal soft-start is provided!!

3. The ripple appearing in signal chain can easily be estimated:
Assume 50mVpp ripple at 100kHz. Then, pi-filter, voltage regulator and RC filter at OPamp will suppress the ripple by about 95dB. So, at the supply terminals of OPamp only 1µVpp will be observed.
Now, OPamp OP249 (I guess you mean this one, not OP247?) also provides a PSRR (power supply rejection ratio) of 35dB. This means, that a ripple of about 20nVpp per OPamp appears in signal chain. Means, each OPamp contributes 20nVpp. So, with two OPamps about 40nVpp ripple will appear, resulting in a final SNR (signal to noise ratio) of about 115dB, assuming a maximum signal voltage of 20mVpp.

This rough estimation assumes a perfectly linear working Opamp. Unfortunately, real OPamps are non ideal. So, due to demodulation processes at internal, non linear working sections of OPamp the 100kHz ripple can appear as noise partially convoluted into the audio frequency range. Also, some unwanted offset effects might be observed, so that the 35dB power supply rejection cannot be fully assumed in all cases. This means, that we should not expect a PSRR of fully 35dB, but somewhat less.

We can roughly estimate the real noise level (worst case) by subtracting the 35dB PSRR from the 115dB SNR, yielding a final SNR of about 80dB.

This last estimation is highly conservative, means you will observe a much higher SNR than the estimated 80dB in very most cases. Experience tells, that a SNR of more than about 100dB is probable!

This rough error budget analysis tells you, where the noise can come from. Here, it tells you, that the suggested ripple filtering will show highly acceptable results. Or, by other words, if you observe a much higher ripple at output of your device, then it will have entered the signal chain via a different path very probably, for instance via magnetic coupling into the coils of guitar (single coil, humbucker,...) or else.

Kai