Kaldi Release 20.10

The NVIDIA container image for Kaldi, release 20.10 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 18.04 including Python 3.6
• NVIDIA CUDA 11.1.0 including cuBLAS 11.2.1
• NVIDIA cuDNN 8.0.4
• NVIDIA NCCL 2.7.8 (optimized for NVLink™)
• MLNX_OFED
• OpenMPI 3.1.6
• Nsight Compute 2020.2.0.18
• Nsight Systems 2020.3.4.32
• TensorRT 7.2.1

Driver Requirements

Release 20.10 is based on NVIDIA CUDA 11.1.0, which requires NVIDIA Driver release 455 or later. However, if you are running on Tesla (for example, T4 or any other Tesla board), you may use NVIDIA driver release 418.xx, 440.30, or 450.xx. 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 20.10 supports CUDA compute capability 6.0 and higher. This corresponds to GPUs in the Pascal, Volta, Turing, and Ampere Architecture GPU 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 20.10 is based on Kaldi 49dab837.
• The latest version of NVIDIA CUDA 11.1.0
• The latest version of cuBLAS 11.2.1
• The latest version of NVIDIA cuDNN 8.0.4
• The latest version of Nsight Compute 2020.2.0.18
• The latest version of Nsight Systems 2020.3.4.32
• The latest version of TensorRT 7.2.1
• Ubuntu 18.04 with September 2020 updates.

Packaged scripts

The Kaldi container comes with the following scripts:

• /workspace/nvidia-examples/librispeech/prepare_data.sh which downloads a trained model and data.
• /workspace/nvidia-examples/librispeech/run_benchmark.sh which runs inference on the trained model and data. Setting the environment variable ONLINE=1 will make the benchmark script run in online mode.

NVIDIA Kaldi Container Versions

The following table shows what versions of Ubuntu, CUDA, Kaldi, and TensorRT are supported in each of the NVIDIA containers for Kaldi. For older container versions, refer to the Frameworks Support Matrix.

Known Issues

There are no known issues in this release.