problem when transcribing a new video

rizkiarm / LipNet

Keras implementation of 'LipNet: End-to-End Sentence-level Lipreading'

MIT License

635 stars 226 forks source link

problem when transcribing a new video #28

Closed simonlight closed 6 years ago

simonlight commented 6 years ago

Thank you for your great work! My question is that when I tried a new video and wanted to generate the transcription from the movement of the lip, I only got unrelated words such as [bin/green/c/nine/soon...] Are there some tricks for correctly running the model for an arbitrary video? Thanks in advance.

rizkiarm commented 6 years ago

You need to train the model for the new words that you want to predict.