Contents of this directory is a modified version of AN4 dataset pre-processed and optimized for CNTK end-to-end testing. The data uses the format required by the HTKMLFReader. For details please refer to the documentation. The AN4 dataset is a part of CMU audio databases. This modified version of dataset is distributed under the terms of a AN4 license which can be found in 'AdditionalFiles/AN4.LICENSE.html'
See License.md in the root level folder of the CNTK repository for full license information.
| Data: | Speech data from the CMU Audio Database aka AN4 (http://www.speech.cs.cmu.edu/databases/an4) |
|---|---|
| Purpose: | Showcase how to train feed forward and LSTM networks for speech data |
| Network: | SimpleNetworkBuilder for 2-layer FF, NdlNetworkBuilder for 3-layer LSTM network |
| Training: | Data-parallel 1-Bit SGD with automatic mini batch rescaling (FF) |
| Comments: | There are two config files: FeedForward.cntk and LSTM-NDL_ndl_deprecated.cntk for FF and LSTM training respectively |
The data for this example is already contained in the folder AN4/Data/.
Compile the sources to generate the cntk executable (not required if you downloaded the binaries).
Windows: Add the folder of the cntk executable to your path
(e.g. set PATH=%PATH%;c:/src/cntk/x64/Debug/;)
or prefix the call to the cntk executable with the corresponding folder.
Linux: Add the folder of the cntk executable to your path
(e.g. export PATH=$PATH:$HOME/src/cntk/build/debug/bin/)
or prefix the call to the cntk executable with the corresponding folder.
Run the example from the Speech/Data folder using:
cntk configFile=../Config/FeedForward.cntk
or run from any folder and specify the Data folder as the currentDirectory,
e.g. running from the Speech folder using:
cntk configFile=Config/FeedForward.cntk currentDirectory=Data
The output folder will be created inside Speech/.
The config files define a RootDir variable and sevearal other variables for directories.
The ConfigDir and ModelDir variables define the folders for additional config files and for model files.
These variables will be overwritten when running on the Philly cluster.
It is therefore recommended to generally use ConfigDir and ModelDir in all config files.
To run on CPU set deviceId = -1, to run on GPU set deviceId to "auto" or a specific value >= 0.
The FeedForward.cntk file uses the SimpleNetworkBuilder to create a 2-layer feed forward network with sigmoid nodes and a softmax layer. The LSTM-NDL_ndl_deprecated.cntk file uses the NdlNetworkBuilder and refers to the lstmp-3layer-opt.ndl file. In the ndl file an LSTM component is defined and used to create a 3-layer LSTM network with a softmax layer. Both configuration only define and execute a single training task:
command=speechTrain
The trained models for each epoch are stored in the output models folder.
The 'AdditionalFiles' folder contains the license terms for the AN4 audio database.