Abstract | ||
---|---|---|
Connectionist temporal classification (CTC) has recently shown improved performance and efficiency in automatic speech recognition. One popular decoding implementation is to use a CTC model to predict the phone posteriors at each frame and then perform Viterbi beam search on a modified WFST network. This is still within the traditional frame synchronous decoding framework. In this paper, the peaky... |
Year | DOI | Venue |
---|---|---|
2017 | 10.1109/TASLP.2016.2625459 | IEEE/ACM Transactions on Audio, Speech, and Language Processing |
Keywords | Field | DocType |
Hidden Markov models,Decoding,Acoustics,Lattices,Pragmatics,Speech recognition,Acoustic beams | Pattern recognition,Computer science,Word error rate,Beam search,Keyword spotting,Speech recognition,Redundancy (engineering),Artificial intelligence,Modular design,Decoding methods,Viterbi algorithm,Speedup | Journal |
Volume | Issue | ISSN |
25 | 1 | 2329-9290 |
Citations | PageRank | References |
1 | 0.35 | 18 |
Authors | ||
4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Zhehuai Chen | 1 | 11 | 3.89 |
Yimeng Zhuang | 2 | 15 | 2.54 |
Yanmin Qian | 3 | 295 | 44.44 |
Kai Yu | 4 | 1082 | 90.58 |