Experiment-text



1. Deep Learning-based Speech Recognition

° LSTM RNN-based Acoustic Model

LSTM RNN-based Acoustic Model


° LSTM RNN-based Language Model

LSTM RNN-based Language Model


° Connectionist Temporal Classification

Connectionist Temporal Classification



2. Audio Event Detection

Audio Event Detection



3. Automatic Generation of Video Transcripts

Automatic Generation of Video Transcripts



4. Unlimited Korean Vocabulary Speech Recognition for Client-Server Environments

Unlimited Korean Speech Recognition
                                for Client-Server Environments




Back to top

Spoken Language Lab, Dep. of CSE, Sogang Univ.
35, Baekbeom-ro, Mapo-gu, Seoul, 04107, Korea (Tel) +82-2-715-2715