[Freeswitch-users] Event socket access to conversation audio

Hi, I'm wondering if the FreeSwitch ESL (or DPtools for that matter) provide any mechanism to directly access the audio channels in a conversation? For instance, instead of just recording to a file, I'd like to pass the channel to python, which will perform some realtime processing on the raw audio, before writing it somewhere else. Is this possible, or do I need to modify the core? Thanks!