I have never tried adapting a model and this will be my first time, I have around 45gb of audio files of about 1400 hours. The specifications of my PC are intel core i7-3770CPU@3.40Ghz*8 Disk: 1TB Hdd and 4GB of RAM, will it be good enough to adapt the default acoustic model?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
For 1400 hours it is better to train from scratch. 4gb of memory is pretty low, something like 16 at least better 64 You also need a gpu card for modern algorithms at least gtx1080.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Thanks for your reply. I will prefer to adapt my model since I have less amount of time. Do you think it would be possible to adapt the model with 1400hours of audio file? or should I reduce the size of the dataset. Thanks
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Thanks for your reply. I will prefer to adapt my model since I have less amount of time. Do you think it would be possible to adapt the model with 1400hours of audio file?
It is possible to adapt but accuracy will not be the best one.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Considering I have a good machine to train the model on, how long will it take for me to train a new model? is it possible to train a model within 10days for 1400hours of speech?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hello. I have downloaded the mozilla common voice english dataset and I have converted the mp3 files to cmusphinx compatible format. I am currently adapting the PTM model to the mozilla common voice dataset. What do you think? how will it affect the accuracy of the model? I am currently adapting on a intel core i7-3770CPU@3.40Ghz*8 Disk: 1TB Hdd and 16GB of RAM. The total size of the dataset is around 50GB and has around 1400 hours of recording in it. Can the model adapt within 5-7 days?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I have never tried adapting a model and this will be my first time, I have around 45gb of audio files of about 1400 hours. The specifications of my PC are intel core i7-3770CPU@3.40Ghz*8 Disk: 1TB Hdd and 4GB of RAM, will it be good enough to adapt the default acoustic model?
For 1400 hours it is better to train from scratch. 4gb of memory is pretty low, something like 16 at least better 64 You also need a gpu card for modern algorithms at least gtx1080.
Thanks for your reply. I will prefer to adapt my model since I have less amount of time. Do you think it would be possible to adapt the model with 1400hours of audio file? or should I reduce the size of the dataset. Thanks
It is possible to adapt but accuracy will not be the best one.
Considering I have a good machine to train the model on, how long will it take for me to train a new model? is it possible to train a model within 10days for 1400hours of speech?
Yes, it is perfectly possible on machine like above.
Hello. I have downloaded the mozilla common voice english dataset and I have converted the mp3 files to cmusphinx compatible format. I am currently adapting the PTM model to the mozilla common voice dataset. What do you think? how will it affect the accuracy of the model? I am currently adapting on a intel core i7-3770CPU@3.40Ghz*8 Disk: 1TB Hdd and 16GB of RAM. The total size of the dataset is around 50GB and has around 1400 hours of recording in it. Can the model adapt within 5-7 days?