Xinjian Li

allosaurus. Allosaurus is a pretrained universal phone recognizer for more than 2000 languages

transphone. phoneme tokenizer and grapheme-to-phoneme model for 8k languages

alqalign. multilingual speech aligner

asr2k. asr2k

ucla-phonetic-corpus. Dataset of ICASSP 2021 MULTILINGUAL PHONETIC DATASET FOR LOW RESOURCE SPEECH RECOGNITION

kaldi-cmake. create CMakeLists.txt for kaldi

pytensor. A numpy deep learning framework

phonepiece. phone inventory library

epitran. A tool for transcribing orthographic text as IPA (International Phonetic Alphabet)

tflstm2np. implement inference logic of tensorflow stack_bidirectional_dynamic_rnn with numpy

newlang-tech. A guide to building language technology in new languages.