The whisper.cpp project ports OpenAI’s Whisper speech recognition model to C and C++.
Open source
24 alternativeswav2letter
The wav2letter project provides an automatic speech recognition toolkit from Facebook AI Research.
About wav2letter
wav2letter is a GitHub repository from Facebook AI Research that provides an automatic speech recognition toolkit. It includes recipes that reproduce several published speech recognition research papers, along with pre-trained models. Data preparation for training and evaluation is documented in a data directory.
The project has been consolidated into Flashlight, and the README states that future wav2letter development will take place there. Reproducing results exactly requires Flashlight version 0.3.2 or earlier. Building the recipes requires installing Flashlight from its 0.3 branch with the ASR application, and the README mentions ArrayFire as a dependency. Users who need the older pre-consolidation version are directed to the wav2letter v0.2 release.
Key features
- Automatic speech recognition toolkit from Facebook AI Research
- Recipes to reproduce five listed speech recognition research papers
- Pre-trained speech recognition models included with the recipes
- Data preparation instructions in a dedicated data directory
- CMake-based build process that depends on Flashlight and ArrayFire
- Recipes covering online, lexicon-free, self-training and sequence-to-sequence recognition
Best for: Speech recognition researchers and developers who want to reproduce the published wav2letter recipes or build on them.
Good for
- Includes recipes that reproduce specific published speech recognition papers
- Provides pre-trained models alongside the recipes
- Lists community channels, including a Facebook group and a Google group
Watch out for
- Development has moved to Flashlight, so this repository is no longer the main development location
- Exact reproduction of results requires Flashlight version 0.3.2 or earlier
- Building requires the Flashlight 0.3 branch and its ASR application, plus ArrayFire
wav2letter at a glance
- Category
- AI & Machine Learning › AI Voice & Audio
- Type
- Open-source project
- Pricing
- Not confirmed. Check the site
- Open source
- Not stated
- Language
- English
- Last checked
- Oct 6, 2026 · Online when checked
- Listed
- Oct 5, 2026
wav2letter alternatives
Tools and sites you could use instead, ranked by how closely they match.
NeMo Parakeet ASR Models attain strong speech recognition accuracy while being efficient for inference. Available in CTC and RNN-Transducer variants.
Superwhisper turns speech into polished text for use in apps such as Slack and Gmail, with recognition options listed for four operating systems.
AudioCraft provides research code for generating music and sound effects and for audio compression.
Kaldi provides an open toolkit for speech recognition and speech processing research.
Text-to-Audio Generation with Latent Diffusion Models - Speech Research.
Popular in AI Voice & Audio
Well-known listings from the same topic.
Google Illuminateilluminate.google.comAlteredaltered.aiELSA Speakelsaspeak.comKreadoAIkreadoai.comPodwisepodwise.aiAddsubtitleaddsubtitle.ai
wav2letter FAQ
Where does future wav2letter development take place?
The README states that wav2letter has been consolidated into Flashlight in the ASR application and that future development will occur in Flashlight.
Which Flashlight version is needed for exact reproduction of the recipe results?
The README states that all results reproduction must use Flashlight 0.3.2 or earlier for exact reproducibility.
Where can I find data preparation steps for training and evaluation?
Data preparation for training and evaluation is described in the data directory of the repository.
Is something wrong or out of date?
Tell us and an editor will check it.
Own this website?
Claim this listing
Claim the listing and suggest updates yourself. An editor checks every claim first.