# Decoders
we borrow a lot of code from Espnet Apache 2.0  (http://www.apache.org/licenses/LICENSE-2.0)

## Reference
### CTC Prefix Beam Search
* [Sequence Modeling With CTC](https://distill.pub/2017/ctc/)
* [First-Pass Large Vocabulary Continuous Speech Recognition using Bi-Directional Recurrent DNNs](https://arxiv.org/pdf/1408.2873.pdf)

### CTC Prefix Score & Join CTC/ATT One-passing Decoding
* [Hybrid CTC/Attention Architecture for End-to-End Speech Recognition](http://www.ifp.illinois.edu/speech/speech_web_lg/slides/2019/watanabe_hybridCTCAttention_2017.pdf)
* [Vectorized Beam Search for CTC-Attention-based Speech Recognition](https://www.isca-speech.org/archive/pdfs/interspeech_2019/seki19b_interspeech.pdf)

### Streaming Join CTC/ATT Beam Search
* [STREAMING TRANSFORMER ASR WITH BLOCKWISE SYNCHRONOUS BEAM SEARCH](https://arxiv.org/abs/2006.14941)