Sreenivasa Rao K

Pradeep Rangan, Sreenivasa Rao K

Manner of articulation detection using deep neural networks require a priori knowledge of the attribute discriminative features or the decent phoneme alignments. However generating an appropriate phoneme alignment is complex and its performance depends on the choice of optimal number of senones, Gaussians, etc. In the first part of our work, we exploit the manner of articulation detection using connectionist temporal classification (CTC) which doesn't need any phoneme alignment. Later we modify the state-of-the-art character based posteriors generated by CTC using the manner of articulation CTC detector. Beam search decoding is performed on the modified posteriors and it's impact on open source datasets such as AN4 and LibriSpeech is observed.

Pradeep R, Sreenivasa Rao K

Conventionally, the manner of articulations in speech signal are derived using discriminative signal processing techniques or deep learning approaches. However, training such complex systems involves feature extraction, phoneme force alignment and deep neural network training. In our work, we initially detect the manner of articulations without phoneme alignment using an end-to-end manner of articulation modeling based on connectionist temporal classification (CTC). The manner of articulation knowledge is deployed in the conventional character CTC path to regenerate the new character CTC path. The modified manner based character CTC is evaluated on open source speech datasets such as AN4, LibriSpeech and TEDLIUM-2 and it outperforms over the baseline character CTC.