https://tools.ietf.org/html/rfc3960 has next part:
Early media refers to media (e.g., audio and video) that is exchanged
   before a particular session is accepted by the called user.  Within a
   dialog, early media occurs from the moment the initial INVITE is sent
   until the User Agent Server (UAS) generates a final response.  It may
   be unidirectional or bidirectional, and can be generated by the
   caller, the callee, or both. Typical examples of early media   generated by 
   the callee are ringing tone and announcements (e.g.,
   queuing status).  Early media generated by the caller typically
   consists of voice commands or dual tone multi-frequency (DTMF) tones
   to drive interactive voice response (IVR) systems.

But I know that in FreeSWITCH in preanswer state I can`t capture DTMF. Why? 
This part of RFC is conflicting with some other standard? It can be very intresting feature.