Apple SpeechAnalyzer API
Hacker ApprovedRanked 6 of 25 speech recognition models by recommenders
Recommended as a fast, accurate on-device speech engine that ships free with macOS; critics note limited language support and missing language autodetection.
Recommended as a fast, accurate on-device speech engine that ships free with macOS; critics note limited language support and missing language autodetection.
29 people gave an opinion
- recommend it
- 25
- criticize it
- 4
- approval share
- 86%
Recommended in 4 of 13 months
- November 2025: recommended
- December 2025: recommended
- January 2026: not recommended
- February 2026: not recommended
- March 2026: not recommended
- April 2026: not recommended
- May 2026: not recommended
- June 2026: not recommended
- July 2026: recommended
- August 2026: not recommended
- September 2026: not recommended
- October 2026: recommended
- November 2026: not recommended (month in progress)
Owned for 3 months
Median of the 3 owners who said how long they have had it.
Each square is one year.
Why it carries the badge
- 25 people recommend itNeeds 10 or more, met.
- 6.3 recommenders for every criticNeeds 2 to 1 or better, met.
- Recommended in 4 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
More accurate than alternatives
“the new API cuts WER 3.5-4x vs the old one (2.12% vs 9.02% on test-clean)”
Comment from 13 Jul 2026 in “Apple's new SpeechAnalyzer API, benchmarked against Whisper and its predecessor” “I noticed better accuracy after switching my app to SpeechAnalyzer”
Comment from 13 Jul 2026 in “Apple's new SpeechAnalyzer API, benchmarked against Whisper and its predecessor” “the most accurate on-device speech engine we tested”
Comment from 13 Jul 2026 in “Apple's new SpeechAnalyzer API, benchmarked against Whisper and its predecessor” - Comment from 13 Jul 2026 in “Apple's new SpeechAnalyzer API, benchmarked against Whisper and its predecessor”
- Comment from 14 Jul 2026 in “Apple's new SpeechAnalyzer API, benchmarked against Whisper and its predecessor”
Fast, transcribes long audio quickly
“it was faster than both small/large Whisper models I compared it to”
Comment from 14 Jul 2026 in “Apple's new SpeechAnalyzer API, benchmarked against Whisper and its predecessor” “SA does indeed run faster and anecdotally more accurately”
Comment from 13 Jul 2026 in “Apple's new SpeechAnalyzer API, benchmarked against Whisper and its predecessor” “In one minute it chews through one hour of audio.”
Comment from 13 Jul 2026 in “Apple's new SpeechAnalyzer API, benchmarked against Whisper and its predecessor” Live results as you talk
“SpeechAnalyzer supports streaming, so you can see what it's hearing from you as you talk. A massive UX improvement.”
Comment from 14 Jul 2026 in “Apple's new SpeechAnalyzer API, benchmarked against Whisper and its predecessor” “it was not only fast and accurate enough, but you also see live results as you talk”
Comment from 14 Jul 2026 in “Apple's new SpeechAnalyzer API, benchmarked against Whisper and its predecessor” Bundled in macOS, always available
“The appeal is that users only have to download it once across all apps that use it.”
Comment from 13 Jul 2026 in “Apple's new SpeechAnalyzer API, benchmarked against Whisper and its predecessor” “SpeechAnalyzer instead comes bundled in macOS and is always available.”
Comment from 14 Jul 2026 in “Apple's new SpeechAnalyzer API, benchmarked against Whisper and its predecessor”
What people warn about
Each mark is one comment
Few languages, no autodetection
“Finally. I'd be delighted though if they actually implemented language autodetection”
Comment from 13 Jul 2026 in “Apple's new SpeechAnalyzer API, benchmarked against Whisper and its predecessor” “it will support like 8 languages and leave the rest hanging for 10 years”
Comment from 14 Jul 2026 in “Apple's new SpeechAnalyzer API, benchmarked against Whisper and its predecessor”
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 |
| Apple SpeechAnalyzer APIThis page | 25 | 4 | 86% | 4 | 3 mo3 owners | Approved |
What people said
28 of the 32 comments that mention it, newest first. Each links to the original comment.
“use the excellent SpeechAnalyzer API on macOS”
“it will support like 8 languages and leave the rest hanging for 10 years”
“it was faster than both small/large Whisper models I compared it to”
“it was not only fast and accurate enough, but you also see live results as you talk”
“SpeechAnalyzer instead comes bundled in macOS and is always available.”
“it’s feels so much more accurate”
“I didn't expect Apple's SpeechAnalyzer to see such an impressive improvement.”
“SpeechAnalyzer supports streaming, so you can see what it's hearing from you as you talk. A massive UX improvement.”
“the most accurate on-device speech engine we tested”
“SA does indeed run faster and anecdotally more accurately”
“SpeechAnalyzer almost always gets it.”
“IF you ALREADY HAVE a MODERN Mac (and trust apple) this is your best option”
“I noticed better accuracy after switching my app to SpeechAnalyzer”
“Finally. I'd be delighted though if they actually implemented language autodetection”
“definitely worried that it's going to 'correct' technical jargon to more common words”
“This is useless test and benchmark”
“The appeal is that users only have to download it once across all apps that use it.”
“the point is to be an on device (and thus local, privacy preserving) option”
“In one minute it chews through one hour of audio.”
“and Apple's VTT writes "T Max".”
“In which world 98% accuracy is not usable?”
“substantially faster and only slightly worse”
“Apple's isn't even usable.”
“the new API cuts WER 3.5-4x vs the old one (2.12% vs 9.02% on test-clean)”
“a powerful combination for offline transcription”
“Are you using to for speech-to-text/text-to-speech?”
“uses both the new SpeechAnalyzer and on device LLM and it has met or exceeded my expectations”
“seems to be better on battery usage”
Counts come from 1,795,640 Hacker News comments, November 2025 to November 2026. How we count. Report a mistake.