Skip to content
SPECIFICTION
Get listed

Kaldi

kaldi-asr.org

Free plan WebWindowsmacOSLinuxAndroidSelf-hosted
Visit kaldi-asr.org
Spec score 8.7 #12 of 32 in Speech Recognition Software

What it is

Kaldi is a free, open-source C++ toolkit for speech recognition aimed at researchers and professionals. Its code and scripts cover methods including linear transforms, MMI, boosted MMI, MCE discriminative training and feature-space discriminative training. The toolkit integrates finite-state transducers through OpenFst and includes a matrix library that wraps standard BLAS and LAPACK routines. Example recipes support building speech recognition systems, and downloadable model archives cover speech recognition, speaker identification, speech activity detection, diarization and language modeling. The models may be used for any purpose. Kaldi is released under Apache License 2.0, which permits commercial use and redistribution. It runs on Linux distributions and Mac OS X, supports Android cross-compilation, and can be compiled to WebAssembly for browser execution. Windows compatibility is not actively maintained, and example scripts do not work there. Users obtain the current development version by cloning the GitHub repository and following its INSTALL file. Documentation is expert-oriented; the project directs help requests to its mailing list and bug reports or feature requests to GitHub issues.

Spec sheet

Purpose
Kaldi is a C++ toolkit for speech recognition intended for speech recognition researchers and professionals.Read 9 Oct 2026 · kaldi-asr.org
Support
Users can seek help through the kaldi-help mailing list, while bug reports and feature requests belong in GitHub issues.Read 9 Oct 2026 · kaldi-asr.org
License
Kaldi is released under Apache License 2.0, which permits free use and redistribution, including for commercial purposes.Read 9 Oct 2026 · kaldi-asr.org
Founded
2009Read 28 Sep 2026 · kaldi-asr.org
Android and browser builds
The project repository says Kaldi supports cross-compiling for Android and WebAssembly for in-browser execution.Read 9 Oct 2026 · github.com
Android support
Kaldi supports cross-compiling for Android using Android NDK, clang++, and OpenBLAS.Read 9 Oct 2026 · github.com
Audience limit
The project says much of its documentation is written for experts and identifies speech recognition researchers and researchers-in-training as its intended audience.Read 9 Oct 2026 · kaldi-asr.org
Compute environment
Kaldi can run on a single Unix-like machine, while its documentation describes a Linux cluster with shared storage and optional Nvidia GPUs as an ideal environment.Read 9 Oct 2026 · kaldi-asr.org
Download
The project directs users to clone its GitHub repository to obtain the current development version, which may include unfinished or experimental features.Read 9 Oct 2026 · kaldi-asr.org
External dependencies
Installation and example scripts use tools including Git, bash, Perl, awk, grep, and make; some example scripts use SRILM, which the documentation says is not free for commercial use.Read 9 Oct 2026 · kaldi-asr.org
Finite-state transducers
Kaldi compiles against the OpenFst toolkit and integrates finite-state transducers into its code.Read 9 Oct 2026 · kaldi-asr.org
FST integration
Kaldi integrates with Finite State Transducers and compiles against the OpenFst toolkit as a library.Read 9 Oct 2026 · kaldi-asr.org
Installation
Users download the current source by cloning the Kaldi GitHub repository and install it using the instructions in the repository's INSTALL file.Read 9 Oct 2026 · kaldi-asr.org
Intended audience
The project says its documentation can be expert-oriented and describes speech recognition researchers and researchers-in-training as its intended audience.Read 9 Oct 2026 · kaldi-asr.org
Linear algebra
Kaldi includes a matrix library that wraps standard BLAS and LAPACK routines.Read 9 Oct 2026 · kaldi-asr.org
Linux support
Kaldi runs on any Linux distribution and is best tested on Debian and Red Hat Linux.Read 9 Oct 2026 · kaldi-asr.org
macOS support
Kaldi runs on Mac OS X.Read 9 Oct 2026 · kaldi-asr.org
Model use
The models page states that its downloadable models may be used for any purpose.Read 9 Oct 2026 · kaldi-asr.org
Modeling recipes
The project provides example recipes for building speech recognition systems and includes models for ASR, speaker identification, speech activity detection, and diarization.Read 9 Oct 2026 · kaldi-asr.org
Models
The site offers downloadable model archives for tasks including speech recognition, speaker identification, speech activity detection, diarization, and language modeling.Read 9 Oct 2026 · kaldi-asr.org
Operating systems
Kaldi is best tested on Debian and Red Hat Linux and can run on Linux distributions, Cygwin, or Mac OS X; Windows compatibility is not actively maintained and example scripts do not work there.Read 9 Oct 2026 · kaldi-asr.org
Project governance
Kaldi describes itself as a community project with no single legal owner and says contributors license their work directly under Apache terms.Read 9 Oct 2026 · kaldi-asr.org
Security and patents
The legal page says Kaldi's use does not protect users against third-party patent claims, including claims by contributors in circumstances described there.Read 9 Oct 2026 · kaldi-asr.org
Speech recognition methods
The project lists code and scripts for standard techniques including linear transforms, MMI, boosted MMI, MCE discriminative training, and feature-space discriminative training.Read 9 Oct 2026 · kaldi-asr.org
Windows limit
Kaldi has been compiled on Windows, but its example scripts do not work there and Windows compatibility is not actively maintained.Read 9 Oct 2026 · kaldi-asr.org

Plans and prices

Kaldi Free Apache License 2.0 · use and redistribution permitted, including commercially Read 9 Oct 2026 · kaldi-asr.org

Every Kaldi plan and price →

Strong points and limits

Strong points

  • Apache License 2.0 permits commercial use.
  • Includes recipes and downloadable model archives.
  • Integrates OpenFst and BLAS/LAPACK routines.
  • Supports Linux, macOS, Android and WebAssembly.

Limits

  • Windows compatibility is not actively maintained.
  • Example scripts do not work on Windows.
  • Documentation is expert-oriented.
  • Some example scripts use SRILM, which is not free for commercial use.

Who it suits

Kaldi is aimed at speech recognition researchers, professionals and researchers-in-training who can work with an expert-oriented toolkit. It is less suited to users seeking maintained Windows compatibility or a beginner-focused guide.

Verdict

Kaldi offers a substantial research toolkit with permissive licensing and model archives for several speech-related tasks. Its expert-oriented documentation and limited Windows support are important considerations.

Where Specifiction places it

Alternatives to KaldiAll alternatives →