Kaldi Release 21.12

The NVIDIA container image for Kaldi, release 21.12 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 20.04 including Python 3.8
NVIDIA CUDA 11.5.0
cuBLAS 11.7.3.1
NVIDIA cuDNN 8.3.1.22
NVIDIA NCCL 2.11.4 (optimized for NVLink™)
rdma-core 36.0
OpenMPI 4.1.1+
OpenUCX 1.11.0rc1
GDRCopy 2.3
NVIDIA HPC-X 2.9
Nsight Systems 2021.3.2.4
TensorRT 8.2.1.8
SHARP 2.5
DALI 1.8
DLProf 1.8.0

Driver Requirements

Release 21.12 is based on NVIDIA CUDA 11.5.0, which requires NVIDIA Driver release 495 or later. However, if you are running on a Data Center GPU (for example, T4 or any other Tesla board), you may use NVIDIA driver release 418.40 (or later R418), 440.33 (or later R440), 450.51 (or later R450), 460.27 (or later R460), or 470.57 (or later R470). 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 and NVIDIA CUDA and Drivers Support.

GPU Requirements

Release 21.12 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 21.12 is based on Kaldi c6772ba4.
Ubuntu 20.04 with November 2021 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.

Container Version	Ubuntu	CUDA Toolkit	Kaldi	TensorRT
21.12	20.04	NVIDIA CUDA 11.5.0	5.5 up to commit	TensorRT 8.2.1.8
21.11		NVIDIA CUDA 11.5.0	5.5 up to commit c6772ba4	TensorRT 8.0.3.4
21.10		NVIDIA CUDA 11.4.2 with cuBLAS 11.6.5.2	5.5 up to commit 2897564e	TensorRT 8.0.3.4
21.09		NVIDIA CUDA 11.4.2	5.5 up to commit 40573bb	TensorRT 8.0.3
21.08		NVIDIA CUDA 11.4.1	5.5 up to commit 851a0a79	TensorRT 8.0.1.6
21.07		NVIDIA CUDA 11.4.0	5.5 up to commit 9bca2777	TensorRT 8.0.1.6
21.06		NVIDIA CUDA 11.3.1	5.5 up to commit b1786aa5	TensorRT 7.2.3.4
21.05		NVIDIA CUDA 11.3.0	5.5 up to commit e92970ae
21.04		NVIDIA CUDA 11.3.0	5.5 up to commit ec554a7c
21.03		NVIDIA CUDA 11.2.1	5.5 up to commit 376281ca	TensorRT 7.2.2.3
21.02		NVIDIA CUDA 11.2.0	5.5 up to commit b78fdcaf	TensorRT 7.2.2.3+cuda11.1.0.024
20.12		NVIDIA CUDA 11.1.1	5.5 up to commit 7905fe5a	TensorRT 7.2.2
20.11	18.04	NVIDIA CUDA 11.1.0	5.5 up to commit edf6963c	TensorRT 7.2.1
20.10		NVIDIA CUDA 11.1.0	5.5 up to commit 49dab837	TensorRT 7.2.1
20.09		NVIDIA CUDA 11.0.3	5.5 up to commit 8372cc6a	TensorRT 7.1.3
20.08		NVIDIA CUDA 11.0.3	5.5 up to commit f88d5a36
20.07		NVIDIA CUDA 11.0.194	5.5 up to commit 67db30cc
20.06		NVIDIA CUDA 11.0.167	5.5 up to commit 2c7e78f	TensorRT 7.1.2
20.03		NVIDIA CUDA 10.2.89	5.5 up to commit da71f301	TensorRT 7.0.0
20.02 20.01			5.5	TensorRT 7.0.0
19.12 19.11				TensorRT 6.0.1
19.10		NVIDIA CUDA 10.1.243
19.09
19.08				TensorRT 5.1.5

Known Issues

None.