研究キーワード (6件):
the result is arbitrary filtered version of sources which dramatically decrease the performance of ASR. How to eliminate the artificial filtering effect while separating sources is my research task.
, it is difficult to get pure speech signal when interference or reflection exist. BSS is a technique to recover the sources from mixtures without any prior information. When applied BSS to speech case
, Pure speech signal is important for Automatic Speech Recognition (ASR) system. But in daily life
, the result is arbitrary filtered version of sources which dramatically decrease the performance of ASR. How to eliminate the artificial filtering effect while separating sources is my research task.
, it is difficult to get pure speech signal when interference or reflection exist. BSS is a technique to recover the sources from mixtures without any prior information. When applied BSS to speech case
, Pure speech signal is important for Automatic Speech Recognition (ASR) system. But in daily life