VoxForge
VoxForge
Hello everyone,
I have recently started learning HTK. I am not an expert in linguistics or speech processing.I am doing a project for lip synchronization in animation. For that I need to do a phonetic transcription and syllabification of a speech file. My input is a .wav file in which a single speaker speaks something. I basically want to learn how to do a speaker dependent phonetic transcription and also it is free alignment not forced alignment. I would be grateful if someone could help me through this. Thank you very much.
--- (Edited on 3/26/2011 7:50 am [GMT-0500] by ) ---
> I basically want to learn how to do a speaker dependent phonetic transcription
For Julius, there are settings you can specify to tell it to output phonetic transcription. For example, when running Julius in server mode, you can tell it to output only the phones using the outcode parameter:
-outcode [W][L][P][S][C][w][l][p][s]
(Only for Server Module Mode) Switch which symbols of recognized
words to be sent to client. Specify 'W' for output symbol, 'L'
for N-gram entry, 'P' for phoneme sequence, 'S' for score, and
'C' for confidence score, respectively. Capital letters are for
the second pass (final result), and small letters are for
results of the first pass. For example, if you want to send
only the output symbols and phone sequences as a recognition
result to a client, specify "-outcode WP".
There should be something similar in HTK (though no server mode)
--- (Edited on 3/30/2011 10:33 am [GMT-0400] by kmaclean) ---
the transcription label files should have the timing information as well. the start and end of each phoneme..does the output of julius is in that format?? thank you
--- (Edited on 4/7/2011 4:26 am [GMT-0500] by Visitor) ---