Task-aware deep bottleneck features for spoken language identification

  title={Task-aware deep bottleneck features for spoken language identification},
  author={Bing Jiang and Yan Song and Si Wei and Ian Vince McLoughlin and Li-Rong Dai},
Recently, deep bottleneck features (DBF) extracted from a deep neural network (DNN) containing a narrow bottleneck layer, have been applied for language identification (LID), and yield significant performance improvement over state-of-the-art methods on NIST LRE 2009. However, the DNN is trained using a large corpus of specific language which is not directly related to the LID task. More recently, lattice based discriminative training methods for extracting more targeted DBF were proposed for… CONTINUE READING
6 Citations
19 References
Similar Papers


Publications referenced by this paper.
Showing 1-10 of 19 references

I-vector representation based on bottleneck features for language identification

  • Y. Song, B. Jiang, Y. Bao, S. Wei, L.-R. Dai
  • Electronics Letters, vol. 49, no. 24, pp. 1569…
  • 2013
Highly Influential
7 Excerpts

Similar Papers

Loading similar papers…