Results on NIST language recognition evaluation 2009 (LRE09) show significant improvements over state-of-the-art systems. To evaluate the effectiveness of this, we design two acoustic models, termed DBF-TV and parallel DBF-TV (PDBF-TV), using a DBF based i-vector representation for each speech utterance. We show that DBFs can form a low-dimensional compact representation of the original inputs with a powerful descriptive and discriminative capability. To address this, we propose using Deep Bottleneck Features (DBF) for spoken LID, motivated by the success of Deep Neural Networks (DNN) in speech recognition.
Furthermore they may be susceptible to the variations caused by different speakers, specific content of the speech segments, and background noise. With the hypothesis that language information is weak and represented only latently in speech, and is largely dependent on the statistical properties of the speech content, existing representations may be insufficient. Although advances in machine learning have led to significant improvements, LID performance is still lacking, especially for short duration speech utterances. For example, in recent years, representations based on both phonotactic and acoustic features have proven their effectiveness for LID. A key problem in spoken language identification (LID) is to design effective representations which are specific to language information.
0 kommentar(er)
