Hello,
I would like to do a recognize speech in French language for an IVR. Would it be possible to do an example for “Hello world”? Because I don’t understand how to use “dynamicdecoder” or “wfsadecoder”.
I installed Bavieca and tools have been created in: “bavieca-code/bin/x86_64-Linux/”.
I read the website bavieca.org, I create files for: “phonetic symbol set”, “hypothesis”, “pronunciation lexicon” and others but I don’t understand how to use for “dynamicdecoder”.
Can you help me?
Best regards.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
To do speech recognition in French you first need speech data to train
a set of acoustic models for French. Bavieca does provide the tools to
do the acoustic modelling, but you need to have the corpus of
transcribed speech. That means that you need a series of utterances
with corresponding word-level transcriptions. Typically hours of
speech are needed, and the speech needs to be from different speakers.
Do you have that kind of data?
Hello,
I would like to do a recognize speech in French language for an IVR. Would
it be possible to do an example for “Hello world”? Because I don’t
understand how to use “dynamicdecoder” or “wfsadecoder”.
I installed Bavieca and tools have been created in:
“bavieca-code/bin/x86_64-Linux/”.
I read the website bavieca.org, I create files for: “phonetic symbol set”,
“hypothesis”, “pronunciation lexicon” and others but I don’t understand how
to use for “dynamicdecoder”.
Can you help me?
Best regards.
To do speech recognition in French you first need speech data to train
a set of acoustic models for French. Bavieca does provide the tools to
do the acoustic modelling, but you need to have the corpus of
transcribed speech. That means that you need a series of utterances
with corresponding word-level transcriptions. Typically hours of
speech are needed, and the speech needs to be from different speakers.
Do you have that kind of data?
Daniel
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Thank you for your attention.
I recover the following files: frenchWords62K.dic, frenchWords62K.DMP that I have recovered from a Sphinx project. From frenchWords62K.dic file, I create a file for the phonetic symbol set : phoneticFrench.
Are they compatible with Bavieca ?
Mickael
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hello,
I would like to do a recognize speech in French language for an IVR. Would it be possible to do an example for “Hello world”? Because I don’t understand how to use “dynamicdecoder” or “wfsadecoder”.
I installed Bavieca and tools have been created in: “bavieca-code/bin/x86_64-Linux/”.
I read the website bavieca.org, I create files for: “phonetic symbol set”, “hypothesis”, “pronunciation lexicon” and others but I don’t understand how to use for “dynamicdecoder”.
Can you help me?
Best regards.
Hello Mickael,
To do speech recognition in French you first need speech data to train
a set of acoustic models for French. Bavieca does provide the tools to
do the acoustic modelling, but you need to have the corpus of
transcribed speech. That means that you need a series of utterances
with corresponding word-level transcriptions. Typically hours of
speech are needed, and the speech needs to be from different speakers.
Do you have that kind of data?
Daniel
On 3/18/13, Mickael micka59@users.sf.net wrote:
Hello Mickael,
To do speech recognition in French you first need speech data to train
a set of acoustic models for French. Bavieca does provide the tools to
do the acoustic modelling, but you need to have the corpus of
transcribed speech. That means that you need a series of utterances
with corresponding word-level transcriptions. Typically hours of
speech are needed, and the speech needs to be from different speakers.
Do you have that kind of data?
Daniel
Hello Daniel,
Thank you for your attention.
I recover the following files: frenchWords62K.dic, frenchWords62K.DMP that I have recovered from a Sphinx project. From frenchWords62K.dic file, I create a file for the phonetic symbol set : phoneticFrench.
Are they compatible with Bavieca ?
Mickael
Mickael, I'm not familiar with those files. Please take a look at the file formats supported by Bavieca: http://www.bavieca.org/tools.html