Sequence-to-Sequence Learning for Romanian Grapheme-to-Phoneme Conversion

This notebook can be used to replicate the results reported in the paper:

Adriana Stan, ”Input Encoding for Sequence-to-Sequence Learning of Romanian Grapheme-to-Phoneme Conversion”, In Proceedings of the 10th Conference on Speech Technology and Human-Computer Dialogue, Timișoara, Romania, 2019, https://sped.pub.ro/

The notebook is based on Keras's tutorial on sequence-to-sequence learning: https://keras.io/examples/lstm_seq2seq/

The MaRePhoR Romanian phonetic transcription dictionary is available here: https://speech.utcluj.ro/marephor/

Questions? adriana.stan@com.utcluj.ro

License: The work is under a GNU General Public License v3.0