--- title: Kaldi ASR Reviews meta_title: 'Kaldi ASR Reviews 2026: Details, Pricing, & Features | G2' meta_description: Filter 21 reviews by the users' company size, role or industry to find out how Kaldi ASR works for a business like yours. aggregate_rating: rating_value: 4.1 review_count: 21 scale: '5' date_modified: '2026-05-05' parent_category: name: Deep Learning url: https://www.g2.com/categories/deep-learning --- # Kaldi ASR Reviews **Vendor:** Slashdot Media **Category:** [Voice Recognition Software](https://www.g2.com/categories/voice-recognition) **Average Rating:** 4.1/5.0 **Total Reviews:** 21 ## About Kaldi ASR Kaldi is an automatic speech recognition toolkit that supports linear transforms, MMI, boosted MMI and MCE discriminative training, feature-space discriminative training, and deep neural networks. ## Kaldi ASR Reviews ### 1. Speaker Verification using Kaldi Toolkit **Rating:** 4.5/5.0 stars **Reviewed by:** Nagendra K. | Senior Engineer - Data Scientist, Enterprise (> 1000 emp.) **Reviewed Date:** June 22, 2021 **What do you like best about Kaldi ASR?** It is open-sourced and very well-maintained toolkit by the core group of Johns Hopkins University's speech recognition laboratory. We can extract various state-of-art features such as i-vector, x-vector which can be used for various speech-related tasks. For speech-related tasks, we can achieve the state-of-art result. **What do you dislike about Kaldi ASR?** Installation of Kaldi with GPU support is a nightmare for me. **Recommendations to others considering Kaldi ASR:** Before installing, first, read the Kaldi group discussion related to installation steps. **What problems is Kaldi ASR solving and how is that benefiting you?** Mostly I have used Kaldi for the Speech-based Speaker Verification task. Training the Total-Variability matrix and Deep Learning model is much faster than the Matlab-based training. ### 2. Current version of Kaldi is not intuitive or user friendly **Rating:** 1.0/5.0 stars **Reviewed by:** Verified User in Information Technology and Services | Small-Business (50 or fewer emp.) **Reviewed Date:** August 13, 2021 **What do you like best about Kaldi ASR?** The upsides of Kaldi is that once you know it very deeply after a lot of experience, the possibilities become quite endless for customising acoustic models. The user community for Kaldi is quite vast, interactive, and odds are that someone has had the same problem as you if you just know what to look for. There are many useful tools in the utils/ folder, even though they all need thorough customisation for appropriate use for the model building, as the process is inherently data-driven. Kaldi does feel like a massive puzzle, and piecing it together is quite rewarding in a strange, masochistic way. It's great that since it is community-based, there are many pre-existing recipes that are easily customisable for various use cases and that you can contribute with your own recipe. My own holy grail that I always go back to is the Eleanor Chodroff tutorial for building Kaldi acoustic models, since it describes the particular data structure required for the process. **What do you dislike about Kaldi ASR?** Well. There are many issues that i must adress pertaining to Kaldi. This is just some of those things that everyone knows and has accepted, but bottom line is that currently Kaldi is not user friendly or intuitive. While there are a lot of recipes, they are all border-line useless because they all need to be thoroughly customised as the point of creating a custom ASR model is that it is entirely data-driven. There are no explanations as to what the many utilities are or why they must occur in which order. The only way to learn how to use Kaldi is through thorough trial and error. If you try to ask Dan Povey questions on the forum, you will get a passive-agressive response thinly veiled as advice telling you to switch careers and stop doing speech recognition. The entire framework is so un-intuitive that it maketh no sense. Literally any user interface or some more comprehensive and straight forward instruction would be great. What also annoys me is that there are so many fantastic language representation systems with which one can make a great LM, but since Kaldi only works with ARPA format, it disallows any great progress in the quality of ASR in regards to LMs. Another thing is that if you make one mistake, you pretty much have to start all over again. Especially since Kaldi is so data-driven, it is particularly difficult to automate AM building processes which is hindering to company growth if Kaldi is the main tool that is used there. **Recommendations to others considering Kaldi ASR:** Patience, trial and error, and lots of reading is key. It can be a powerful tool and create great models, but starting out can feel quite slow. **What problems is Kaldi ASR solving and how is that benefiting you?** I use Kaldi to build custom ASR models for clients, the benefits were that it was pretty much the only sensible option for ASR building at the time. ### 3. Kaldi is user-friendly tool, which gives us a freedom to explore the things like speech recognition. **Rating:** 5.0/5.0 stars **Reviewed by:** Nadeem P. | Machine Learning Engineer, Mid-Market (51-1000 emp.) **Reviewed Date:** June 26, 2021 **What do you like best about Kaldi ASR?** Language Model creation and FST creation. **What do you dislike about Kaldi ASR?** Lexicon generation requires linguists help if open source lexicon data is not available. **Recommendations to others considering Kaldi ASR:** If someone want to understand the practical experience of ASR working. **What problems is Kaldi ASR solving and how is that benefiting you?** Domain specific ASR. ### 4. I have a great experience using kaldi toolkit . **Rating:** 5.0/5.0 stars **Reviewed by:** Ayush J. | Software developer, Small-Business (50 or fewer emp.) **Reviewed Date:** June 20, 2021 **What do you like best about Kaldi ASR?** Speed, accuracy. It makes the job simpler. Speed was great. All the documentation was there. The instruction was really helpful. There is no other tool like kaldi to implement the speech-to-text conversion. **What do you dislike about Kaldi ASR?** Operating system compatibility. I faced a problem with windows OS. Kaldi was faster in Linux but it was difficult to implement in windows. **Recommendations to others considering Kaldi ASR:** Highly recommend kaldi for speech text conversion. Easy to use and implement. Instruction for implementation was well documented and each step was very clear. **What problems is Kaldi ASR solving and how is that benefiting you?** Speech to text conversion. understanding the speech data using kaldi was awesome. ### 5. Kaldi - a tool for customized and time synchronized ASR **Rating:** 4.0/5.0 stars **Reviewed by:** Verified User in Primary/Secondary Education | Small-Business (50 or fewer emp.) **Reviewed Date:** June 28, 2021 **What do you like best about Kaldi ASR?** It has fst for LM which makes it very flexible and customizable solution to target application domain. It also renders the phoneme time stamps in ctm output, which makes it an ideal solution for time synchronization and confidence score calibration **What do you dislike about Kaldi ASR?** It needs a lots and lots of memory resources to load the bulky acoustic models and the LM graphs. **What problems is Kaldi ASR solving and how is that benefiting you?** Fluency and Pronunciation scoring for kids edutech. We could precisely score the confidence scores for spoken sentence based from deterministic lattice indexing and precise time synchronization. ### 6. kaldi is very well thought and written tool **Rating:** 5.0/5.0 stars **Reviewed by:** Verified User in Information Technology and Services | Small-Business (50 or fewer emp.) **Reviewed Date:** June 23, 2021 **What do you like best about Kaldi ASR?** recipes, stability, and user friendly, Very smart and intelligent people worked for it. Kaldi is an excellent toolkit that continually lead the research in ASR technologies **What do you dislike about Kaldi ASR?** The base code is in c++. In today's time, if it is in python, it would be much more easily accessible to broader people. **What problems is Kaldi ASR solving and how is that benefiting you?** I have trained speech recognition systems with Kaldi. Right now, I use the feature extraction pipeline for training end-to-end ASR systems. ### 7. Kaldi is a helpful tool for speech recognition. **Rating:** 4.0/5.0 stars **Reviewed by:** Verified User in Hospital & Health Care | Small-Business (50 or fewer emp.) **Reviewed Date:** July 03, 2021 **What do you like best about Kaldi ASR?** It is very convenient and useful to convert audio files to structured files. It can be used in many coding languages, including Python and C++. Its automatical process helps save time. **What do you dislike about Kaldi ASR?** The handbook of Kaldi is not clear enough and sometimes you need to google and check to totally understand the meaning of some parameters. **What problems is Kaldi ASR solving and how is that benefiting you?** I was working on the development of a real-time speech recognition AI for medical and healthcare conversations. Kaldi helps me convert audio files to structured file and then I can apply speech recognition models. ### 8. Very useful but limited for use cases **Rating:** 4.0/5.0 stars **Reviewed by:** Verified User in Higher Education | Mid-Market (51-1000 emp.) **Reviewed Date:** June 28, 2021 **What do you like best about Kaldi ASR?** Kaldi tool is very fast and easy to handle. **What do you dislike about Kaldi ASR?** At the initial point, it is tough to learn. If you are learning it alone then it looks tough to use it. **Recommendations to others considering Kaldi ASR:** I would like to recommend it to all those works in the same domain. **What problems is Kaldi ASR solving and how is that benefiting you?** Speaker Recognition using Neural network. This has very nice implementation of NN. ### 9. Kaldi is a very good software for both beginners and advanced speech research. **Rating:** 5.0/5.0 stars **Reviewed by:** Verified User in Computer Software | Small-Business (50 or fewer emp.) **Reviewed Date:** August 18, 2021 **What do you like best about Kaldi ASR?** The features. Like multiple algorithms for feature extraction. Support for many neural architectures. **What do you dislike about Kaldi ASR?** Unless we are masters in C++, its quite difficult to hack into the source code. **What problems is Kaldi ASR solving and how is that benefiting you?** Speech recognition, speaker identification, and language identification. The benefits are it has support for many previously published papers, and it's very easy to reproduce the results. ### 10. useful for all the speech researchers **Rating:** 4.5/5.0 stars **Reviewed by:** Verified User in Computer Software | Small-Business (50 or fewer emp.) **Reviewed Date:** July 03, 2021 **What do you like best about Kaldi ASR?** easy sample script access for building speech based models. **What do you dislike about Kaldi ASR?** It cannot handle end-to-end architecture models. Provision should be provided for those. **Recommendations to others considering Kaldi ASR:** A must for beginners in speech **What problems is Kaldi ASR solving and how is that benefiting you?** primary Industry use cases can be solved using Kaldi. Simple to implement ### 11. As of today Kaldi is awesome tool for speech recognition development **Rating:** 3.5/5.0 stars **Reviewed by:** Verified User in Information Technology and Services | Enterprise (> 1000 emp.) **Reviewed Date:** July 04, 2021 **What do you like best about Kaldi ASR?** WFST, DNN, nnet2 & nnet3, available for live streaming detection and can take it to cloud **What do you dislike about Kaldi ASR?** mono alignment, need alignment-free approach like CTC and need more attention for small footprint memory on device supports. **Recommendations to others considering Kaldi ASR:** Kadi is the best tool for speech recognition using deep neural network training **What problems is Kaldi ASR solving and how is that benefiting you?** For large vocabulary speech recognition... Kaldi supporting a greater level. In terms of accuracy also great. One of the best tool from research to production. ### 12. Kaldi review for someone's use **Rating:** 4.0/5.0 stars **Reviewed by:** Qasim A. | Data Science Engineer, Small-Business (50 or fewer emp.) **Reviewed Date:** June 26, 2021 **What do you like best about Kaldi ASR?** ASR capability and Speaker Diarization in Kaldi **What do you dislike about Kaldi ASR?** Setup and installation is very complicated **What problems is Kaldi ASR solving and how is that benefiting you?** Built an ASR for Urdu language and the benefit I realized in Kaldi is the ability to change anything you want. ### 13. kaldi **Rating:** 5.0/5.0 stars **Reviewed by:** Verified User in Computer Software | Mid-Market (51-1000 emp.) **Reviewed Date:** June 29, 2021 **What do you like best about Kaldi ASR?** kaldi is a reliable speech recognition tool. it performs better than nuance as it has better language models. **What do you dislike about Kaldi ASR?** only downside is nuance has a wider variety of language models for non native speakers **What problems is Kaldi ASR solving and how is that benefiting you?** speech recognition within the warehouse is a huge market especially with the warehouse workers. for kaldi to be able to recognize key words within a noisy environment is huge ### 14. Getting used to kaldi for beginners is a bit tidy . it has low level binaries that are encapsulated **Rating:** 4.0/5.0 stars **Reviewed by:** Verified User in Higher Education | Enterprise (> 1000 emp.) **Reviewed Date:** June 25, 2021 **What do you like best about Kaldi ASR?** Tha kaldi zupport is the best and the development is on going and it has vast algorithms implented and better commented at places. **What do you dislike about Kaldi ASR?** The documentation for certain packages can be improved. **What problems is Kaldi ASR solving and how is that benefiting you?** Using kaldi for indian languages and it has already algorithms tested on big dataset such as WSJ, timit etc. which makes research easier ### 15. I was working with Kaldi almost from last 7years. I found this tool very useful for ASR **Rating:** 4.5/5.0 stars **Reviewed by:** Verified User in Research | Small-Business (50 or fewer emp.) **Reviewed Date:** July 05, 2021 **What do you like best about Kaldi ASR?** User friendly scripts and all other setup, onces you wil learn how to build system using kaldi then this easiest tool. **What do you dislike about Kaldi ASR?** not particluar specific, overall it is good **Recommendations to others considering Kaldi ASR:** You can use this tool for your reserach purpose. **What problems is Kaldi ASR solving and how is that benefiting you?** I was using to build ASR for children speeh. ### 16. I am using Kaldi for transcription and i found this best online and offline transcription tool **Rating:** 5.0/5.0 stars **Reviewed by:** Verified User in Airlines/Aviation | Mid-Market (51-1000 emp.) **Reviewed Date:** June 20, 2021 **What do you like best about Kaldi ASR?** Transcription Accuracy. Also the libraries are very helpful **What do you dislike about Kaldi ASR?** Documentations really lake and naive user cannot get into the informatiln also their is not best platform like stackoverflow to remove the error concern but rather a google group **Recommendations to others considering Kaldi ASR:** Improve documentation and support platform **What problems is Kaldi ASR solving and how is that benefiting you?** Speech to text transcription ### 17. It's good for learning but fails on production at present **Rating:** 1.0/5.0 stars **Reviewed by:** Verified User in Computer Software | Small-Business (50 or fewer emp.) **Reviewed Date:** June 28, 2021 **What do you like best about Kaldi ASR?** Great documentation and many tutorials, the community is also very responsive. **What do you dislike about Kaldi ASR?** The neural network part of kaldi is out-of-dated, the pipe style command line makes it hard to run on Windows system. **What problems is Kaldi ASR solving and how is that benefiting you?** It serves as a bootstrap code for ASR system. ### 18. Kaldi ASR review **Rating:** 3.5/5.0 stars **Reviewed by:** Verified User in Information Technology and Services | Small-Business (50 or fewer emp.) **Reviewed Date:** June 24, 2021 **What do you like best about Kaldi ASR?** Provides high quality ASR results, lots of freedom on how to do things. C++ makes it fast. **What do you dislike about Kaldi ASR?** The framework is difficult to use, and lots of times require an in depth understanding of ASR concepts. **What problems is Kaldi ASR solving and how is that benefiting you?** I'm solving speech-to-text and voice-analytics-related problems. ### 19. Best tool kit to build ASR **Rating:** 5.0/5.0 stars **Reviewed by:** Verified User in Computer Software | Small-Business (50 or fewer emp.) **Reviewed Date:** July 01, 2021 **What do you like best about Kaldi ASR?** Modular structure and highliy reliable and robust **What do you dislike about Kaldi ASR?** Less help and lack of helpful people unclear documentation **Recommendations to others considering Kaldi ASR:** Be persistent you will get results for sure **What problems is Kaldi ASR solving and how is that benefiting you?** ASR ,gives good accuracy and lightweight ,good training speed ### 20. I had used Kaldi during my internship at Nvidia, blazingly fast and the codebase is amazing **Rating:** 5.0/5.0 stars **Reviewed by:** Verified User in Computer Software | Enterprise (> 1000 emp.) **Reviewed Date:** July 01, 2021 **What do you like best about Kaldi ASR?** Kaldi's GPU acceleration and its supported models and algorithms **What do you dislike about Kaldi ASR?** Not quite having anything that I dislike **Recommendations to others considering Kaldi ASR:** Contribute to it if you have chances **What problems is Kaldi ASR solving and how is that benefiting you?** Automated Speech Recognition ### 21. My review on Kaldi **Rating:** 4.5/5.0 stars **Reviewed by:** Kyu H. | Principal Machine Learning Scientist, Computer Software, Small-Business (50 or fewer emp.) **Reviewed Date:** January 04, 2018 **What do you like best about Kaldi ASR?** It's open-sourced, but very well maintained by the core group of Johns Hopkins University's speech recognition laboratory. It has been offering a great suite of tools and libraries for speech scientists to utilize to perform their research experiments with ease as well as for industry practitioners to exploit to build business solutions for customers who are interested in paying for dependable speech recognition solutions. It also supports various neural network architectures that can be useful for any sequential processing. Given that we're in the era of deep learning booming, this tool is a real gem for any machine learning engineer to be able to use to build and evaluate neural network based systems for various tasks. **What do you dislike about Kaldi ASR?** There have been too often changes in file IO interfaces, which make users confused sometimes. **Recommendations to others considering Kaldi ASR:** Kaldi is a great suite of tools and libraries for speech scientists or machine learning engineers who are interested in speech recognition to build and test state-of-the-art speech recognition systems! **What problems is Kaldi ASR solving and how is that benefiting you?** We're building speech recognition solutions using Kaldi for our customers. Kaldi enables us achieve the state-of-the-art performance, even competing with big companies like Google or Microsoft. ## Kaldi ASR Discussions - [What can Kaldi do?](https://www.g2.com/discussions/what-can-kaldi-do) - [How good is Kaldi?](https://www.g2.com/discussions/how-good-is-kaldi) - [Who uses Kaldi?](https://www.g2.com/discussions/who-uses-kaldi) - [Will Kaldi 2.0 be more user friendly, and will it be compatible with other language representation models such as BERT for language modelling?](https://www.g2.com/discussions/will-kaldi-2-0-be-more-user-friendly-and-will-it-be-compatible-with-other-language-representation-models-such-as-bert-for-language-modelling) - 1 upvote - [mjbkkjkjj](https://www.g2.com/discussions/mjbkkjkjj) - [View Kaldi ASR pricing details and edition comparison](https://www.g2.com/products/kaldi-asr/reviews?section=pricing&secure%5Bexpires_at%5D=2026-06-28+02%3A17%3A50+-0500&secure%5Bsession_id%5D=c1f40e93-7334-4e20-a6e6-6889dbfd1a47&secure%5Btoken%5D=db4c521be284bb092f37b7c1662c5034e4d6cc93007061c102190ca59f25c794&format=llm_user) ## Kaldi ASR Features **Deployment & Integration - Voice Recognition** - Installation & setup Ease - Developer API & SDK - Software Integration - Multi-Device Support **Performance Optimization - Voice Recognition** - Accuracy in Noisy Settings - High-Volume Scalability - Environmental Noise Adaptation - Multilingual Voice Recognition - Low-Latency Processing **Security & Compliance - Voice Recognition** - Liveness Detection - Regulatory Compliance - Secure Communication Channels **Advanced AI & Biometric Features - Voice Recognition** - Voice-Based Authentication - Machine Learning & Adaptive Speech Recognition - Speaker Differentiation - Sentiment & Tone Analysis **Agentic AI - Voice Recognition** - Natural Language Interaction ## Top Kaldi ASR Alternatives - [OpenAI Whisper](https://www.g2.com/products/openai-whisper/reviews) - 4.6/5.0 (19 reviews) - [Deepgram](https://www.g2.com/products/deepgram/reviews) - 4.6/5.0 (443 reviews) - [Otter.ai](https://www.g2.com/products/otter-ai/reviews) - 4.4/5.0 (493 reviews)