Our goal is a system for the separation of two speakers during teleconferencing or for hearing aids. To be useful in real time, we want it to work online with as low delay as possible. Proposed approach works in time domain, using attenuation factors and fractional delays between microphone signals to minimize cross-talk, the principle of a fractional delay and sum beamformer. Compared to other approaches this has the advantage that we have lower computational complexity, no system delay and no musical noise like in frequency domain algorithms. We evaluate our approach on convolutive mixtures generated from speech signals taken from the TIMIT data-set using a room impulse response simulator.
(See document for exact affiliation information.)
Click to purchase paper as a non-member or login as an AES member. If your company or school subscribes to the E-Library then switch to the institutional version. If you are not an AES member Join the AES. If you need to check your member status, login to the Member Portal.
