METHOD OF MANAGING A SPEECH MEMORY, AND ASSEMBLY THEREFOR

A method of managing a speech memory (VMEM) and an assembly therefor are disclosed. To enable a plurality of users (U1, U2, U3, U4) to access the speech memory, the latter is divided into user memory areas (S1, S2, S3, S4) and a common memory area (SALL). By adding a portion (S4') of the common memory area (SALL) to one of the user memory areas (S4') or vice versa, the size of this user memory area is adapted to the memory space requirements (Min4) of the user (U4). The capacity of the speech memory (VMEM) is effectively utilized, since, as far as possible, no free memory space is assigned to the users (U1 to U4) which will not be used by them. The assembly can be used, for example, as a voice mailbox in a telecommunications system.