I have tried to use existing acoustic models for telephonic data, but i am getting only about 5% accuracy, so i proceeded with training the acoustic model, however this too doesn't seem to improve the accuracy. I want to develop a speaker independent system for telephonic data(conversation between two speakers). Any help is appreciated. Are there any paid models available for the same?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
In order to get help on the accuracy you need to provide more information about data you want to decode and decoder configuration. You need to provide the audio file you want to recognize, describe software versions you use and provide the decoder output.
The commercial models for telephony data are available, you can contact me at nshmyrev@gmail.com for details.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I have tried to use existing acoustic models for telephonic data, but i am getting only about 5% accuracy, so i proceeded with training the acoustic model, however this too doesn't seem to improve the accuracy. I want to develop a speaker independent system for telephonic data(conversation between two speakers). Any help is appreciated. Are there any paid models available for the same?
In order to get help on the accuracy you need to provide more information about data you want to decode and decoder configuration. You need to provide the audio file you want to recognize, describe software versions you use and provide the decoder output.
The commercial models for telephony data are available, you can contact me at nshmyrev@gmail.com for details.