Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Command line interface for the built-in speech recognition and transcription capabilities in macOS.
| Date | Stars |
|---|---|
| 2026-07-24 | 669 |
| 2026-07-25 | 669 |
| 2026-07-28 | 669 |
| 2026-07-30 | 669 |
| 2026-07-31 | 669 |
| 2026-08-06 | 669 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
[](https://en.wikipedia.org/wiki/MacOS) [](https://opensource.org/licenses/BSD-3-Clause) [](https://en.wikipedia.org/wiki/Objective-C) [](https://github.com/sveinbjornt/hear/releases) [](https://github.com/sveinbjornt/hear/actions) # hear <img src="assets/icon.png" width="168" height="168" alt="hear" style="float: right; margin-left: 20px; margin-bottom: 20px;" align="right"> ### macOS speech recognition via the command line `hear` is a command line interface for the built-in speech recognition capabilities in macOS. This makes it possible to transcribe spoken audio input or entire audio files. Apple's macOS has long shipped with the `say` tool, a command line interface for the operating system's venerable speech synthesis capabilities. As of 10.15 "Catalina", it also ships with highly capable speech recognition for a variety of languages. This project is an attempt to provide a robust native CLI for this functionality since no such tool is provided by Apple. `hear` supports speech recognition and transcription of both live microphone input and audio files in a variety of formats. All formats supported by CoreAudio should work (e.g. WAV, MP3, AIFF, AAC, CAF, ALAC, etc.). See the [man page](https://sveinbjorn.org/files/manpages/hear.1.html) for further details. ## Download <a href="https://www.paypal.com/cgi-bin/webscr?cmd=_s-xclick&hosted_button_id=BDT58J7HYKAEE"><img align="right" src="https://www.paypalobjects.com/WEBSCR-640-20110306-1/en_US/i/btn/btn_donate_LG.gif" ></a> `hear` is free, open source software written in Objective-C. The source code is freely [available](https://github.com/sveinbjornt/hear) under a [BSD license](#bsd-license). **If you find this program useful, please [make a donation](https://sveinbjorn.org/donations).** * **[⬇ Download hear 0.8](https://sveinbjorn.org/files/software/hear.zip)** (~50 KB, ARM/Intel 64-bit, macOS 13 or later, Developer ID signed and notarized by Apple) ## Installation After downloading and expanding the archive, change to the resulting directory and run the following command: ```bash sudo bash install.sh ``` By default, this will install the binary into `/usr/local/bin` and the man page into `/usr/local/share/man/man1/`. ### Install via Homebrew A full installation of [Xcode](https://developer.apple.com/xcode/resources/) is required to build `hear` via Homebrew. ```bash brew tap sveinbjornt/hear https://github.com/sveinbjornt/hear brew install sveinbjornt/hear/hear ``` ## Usage ### Transcribe audio from microphone (or default audio input device) ```bash hear ``` ### From microphone in single line output mode ```bash hear -m ``` ### Transcribe audio file ```bash hear -d -i /path/to/someone_speaking.wav > transcribed_text.txt ``` The `-d` flag specifies that only on-device speech recognition capabilities should be used. Otherwise, data may be sent to Apple servers, which as of writing (2026) have a hard limit of about 500 characters before quitting. For some reason, only those locales supported by Siri work with the `-d` flag. ## Build To build `hear` from source, run the following command from the repository root (requires Xcode command line build tools): ```bash make build_unsigned ``` The resulting binary is created in `products/`. ## Test To run tests on the built command line tool, execute the following command from the repository root: ```bash bash test/test.sh ``` ## Troubleshooting If running the binary leads to an abort signal, try running it by right-clicking it in the Finder and selecting Open. This should automatically open the binary with y
Excerpt of 5,598 characters
Read on GitHub
124
2
2
1
1
1
1
Would you bet a product on this? Bounded 0–100 and slow moving.
matched fp:a6498db706aa2633, topic:speech-recognition, desc:speech recognition, readme:speech recognition