whisper.cpp
Hacker ApprovedRanked 5 of 25 speech recognition models by recommenders
Fast, free, fully local speech-to-text that people run for dictation, subtitles and meetings. Critics say speed depends heavily on hardware, especially without a high-end GPU.
Fast, free, fully local speech-to-text that people run for dictation, subtitles and meetings. Critics say speed depends heavily on hardware, especially without a high-end GPU.
34 people gave an opinion
- recommend it
- 33
- criticize it
- 1
- approval share
- 97%
Recommended in 11 of 13 months
- November 2025: recommended
- December 2025: recommended
- January 2026: recommended
- February 2026: recommended
- March 2026: recommended
- April 2026: recommended
- May 2026: recommended
- June 2026: not recommended
- July 2026: recommended
- August 2026: recommended
- September 2026: recommended
- October 2026: recommended
- November 2026: not recommended (month in progress)
Owned for 2 years
Median of the 4 owners who said how long they have had it.
Each square is one year.
Why it carries the badge
- 33 people recommend itNeeds 10 or more, met.
- 33 recommenders for every criticNeeds 2 to 1 or better, met.
- Recommended in 11 of 13 monthsNeeds 4 or more, met.
The same three rules apply to every product, and the badge can’t be bought. How the badge works
What people say about it
The reasons people give, with the number of comments that give each one. Open a reason to read what they said.
Why people recommend it
Each mark is one comment
Runs locally across platforms
“The moat here is local inference. Whisper.cpp + Metal gives you <500ms latency on M1 with the small model.”
Comment from 17 Feb 2026 in “Show HN: Free Alternative to Wispr Flow, Superwhisper, and Monologue” “A local-only voice to text whisper.cpp transcriber I can globally use while holding ctrl-semicolon.”
Comment from 4 Mar 2026 in “Glaze by Raycast” “tend to just work perfectly, across a whole bunch of different platforms and hardware backends”
Comment from 14 May 2026 in “What's in a GGUF, besides the weights – and what's still missing?” - Comment from 30 Mar 2026 in “Screamer – The fastest free speech to text AI in the world”
- Comment from 14 Nov 2025 in “Show HN: Chirp – Local Windows dictation with ParakeetV3 no executable required”
Fast transcription, even on CPU
“works at 6x realtime transcription speed on my four-core ~2.4GHz laptop, and doesn't really stress CPU or memory”
Comment from 17 Feb 2026 in “Show HN: Free Alternative to Wispr Flow, Superwhisper, and Monologue” “The performance of whisper.cpp allows me to transcribe 1h of meeting in just 20-30s”
Comment from 14 Feb 2026 in “Meeting-Assistant, Local meeting notes assistant and AI analysis in C++” “Whisper.cpp/Faster-whisper are a good bit faster than OpenAI's implementation”
Comment from 7 Nov 2025 in “You should write an agent” - Comment from 19 Jul 2026 in “Transcribe.cpp”
- Comment from 8 Jan 2026 in “Dell's CES 2026 chat was the most pleasingly un-AI briefing I've had in 5 years”
Good output quality
“technical terms are going to be just fine with Whisper”
Comment from 14 Nov 2025 in “Show HN: Chirp – Local Windows dictation with ParakeetV3 no executable required” “it detects very well my very-local spanish modisms”
Comment from 28 Aug 2026 in “Gemini-3.5-Transcribe” “leaps and bounds better in terms of output quality”
Comment from 2 Oct 2026 in “Mozilla shutting down Solo AI website creator” - Comment from 27 Sep 2026 in “Show HN: Jauvex, one app for Claude+Codex+Grok+Jev with two-way voice chat”
Reliable day to day
“tend to just work perfectly, across a whole bunch of different platforms and hardware backends”
Comment from 14 May 2026 in “What's in a GGUF, besides the weights – and what's still missing?” “it just keeps on working without any issues”
Comment from 19 Jul 2026 in “Transcribe.cpp” “has been pretty solid for me”
Comment from 26 May 2026 in “Ask HN: Best Free Local STT?”
What people warn about
Each mark is one comment
Slow on some hardware or workloads
“Whisper gives decent performance as far as speed goes only on a pretty top-end GPU”
Comment from 19 Jul 2026 in “Transcribe.cpp” “whisper.cpp is awesome but even it just can't process a three-hour vid in seconds”
Comment from 10 Sep 2026 in “Why Emacs Consult async searches feel slow and how to speed them up” “yet it completes in a fraction of the time using batching”
Comment from 9 Jan 2026 in “Dell's CES 2026 chat was the most pleasingly un-AI briefing I've had in 5 years” - Comment from 12 Aug 2026 in “llama.cpp”
Other speech recognition models people recommend
The four most recommended other speech recognition models, with the same counts for each, so you can compare like for like. The tick on each bar marks the 2 to 1 line the badge needs.
| Product | Recommend | Criticize | Approval | Months | Owned | Badge |
|---|---|---|---|---|---|---|
| OpenAI Whisper | 164 | 53 | 76% | 12 | — | Approved |
| NVIDIA Parakeet | 144 | 27 | 84% | 12 | — | Approved |
| Mistral Voxtral | 69 | 27 | 72% | 8 | — | Approved |
| Moonshine open-weights STT models | 37 | 13 | 74% | 7 | — | Approved |
| whisper.cppThis page | 33 | 1 | 97% | 11 | 2 yrs4 owners | Approved |
What people said
30 of the 35 comments that mention it, newest first. Each links to the original comment.
“I defaulted to whisper.cpp”
“the insanse community work on Apple Silicon across whisper.cpp/mflux/llama.cpp/MLX/Exo”
“it seems like I made the right move by sticking with whiser.cpp this whole time”
“leaps and bounds better in terms of output quality”
“Whisper.cpp has given better results”
“whisper.cpp is awesome but even it just can't process a three-hour vid in seconds”
“it detects very well my very-local spanish modisms”
“performs 2-3x worse than CPU on my Snapdragon X2 laptop”
“really smoothed out the UX”
“Whisper gives decent performance as far as speed goes only on a pretty top-end GPU”
“whisper.cpp has realtime capability. Been using it for 2 years at this point”
“it just keeps on working without any issues”
“Likewise with simple whisper.cpp”
“has been pretty solid for me”
“tend to just work perfectly, across a whole bunch of different platforms and hardware backends”
“It makes this kind of task actually bearable.”
“Rust + whisper.cpp, Metal GPU acceleration, runs entirely on-device”
“A local-only voice to text whisper.cpp transcriber I can globally use while holding ctrl-semicolon.”
“works at 6x realtime transcription speed on my four-core ~2.4GHz laptop, and doesn't really stress CPU or memory”
“The moat here is local inference. Whisper.cpp + Metal gives you <500ms latency on M1 with the small model.”
“The performance of whisper.cpp allows me to transcribe 1h of meeting in just 20-30s”
“Whisper.cpp is better with transcribing full phrases than with streaming.”
“yet it completes in a fraction of the time using batching”
“It's probably the fastest CPU only version.”
“Cool. I had to revert "vulkan: Replace deprecated VK_EXT_validation_features (llama/17637)" to get it to build on my system.”
“Big shout out to https://github.com/mybigday/whisper.rn and https://huggingface.co/ggerganov/whisper.cpp/tree/main for making this even possible.”
“Some models are useful; using whisper.cpp comes to mind to create subtitles for, for example, family videos”
“technical terms are going to be just fine with Whisper”
“I use it for Linux using whisper CPP and it works great”
“Whisper.cpp/Faster-whisper are a good bit faster than OpenAI's implementation”
Counts come from 1,795,640 Hacker News comments, November 2025 to November 2026. How we count. Report a mistake.