Kaldi Release 19.03

The NVIDIA container image for Kaldi, release 19.03, is available on NGC.

Contents of the Kaldi container

This container image contains the complete source of the version of Kaldi in /opt/kaldi. The container also includes the following:

• Ubuntu 16.04 including Python 3.5
• NVIDIA CUDA 10.1.105 including cuBLAS 10.1.105
• NVIDIA cuDNN 7.5.0
• NVIDIA NCCL 2.4.3 (optimized for NVLink™ )
• OpenMPI 3.1.3
• TensorRT 5.1.2

Driver Requirements

Release 19.03 is based on CUDA 10.1, which requires NVIDIA Driver release 418.xx+. However, if you are running on Tesla (Tesla V100, Tesla P4, Tesla P40, or Tesla P100), you may use NVIDIA driver release 384.111+ or 410. The CUDA driver's compatibility package only supports particular drivers. For a complete list of supported drivers, see the CUDA Application Compatibility topic. For more information, see CUDA Compatibility and Upgrades.

GPU Requirements

Release 19.03 supports CUDA compute capability 6.0 and higher. This corresponds to GPUs in the Pascal, Volta, and Turing families. Specifically, for a list of GPUs that this compute capability corresponds to, see CUDA GPUs. For additional support details, see Deep Learning Frameworks Support Matrix.

Key Features and Enhancements

This Kaldi release includes the following key features and enhancements.

• Kaldi container image version 19.03 is based on Kaldi 5.5.
• Latest version of NVIDIA CUDA 10.1.105 including cuBLAS 10.1.105
• Latest version of NVIDIA cuDNN 7.5.0
• Latest version of NVIDIA NCCL 2.4.3
• Latest version of OpenMPI 3.1.3
• Latest version of TensorRT 5.1.2
• This is the first release for the optimized Kaldi container.
• Ubuntu 16.04 with February 2019 updates

Packaged scripts

The Kaldi container comes with the following scripts:

• /workspace/nvidia-examples/librispeech/prepare.sh which downloads a trained model and data.
• /workspace/nvidia-examples/librispeech/run_benchmark.sh and /workspace/nvidia-examples/librispeech/run_multigpu_benchmark.sh which run inference on the trained model and data.

Known Issues

• If using or upgrading to a 3-part-version driver, for example, a driver that takes the format of xxx.yy.zz, you will receive a Failed to detect NVIDIA driver version. message. This is due to a known bug in the entry point script's parsing of 3-part driver versions. This message is non-fatal and can be ignored. This will be fixed in the 19.04 release.