Superwhisper Review
Local-first dictation with cloud AI modes
macOS Windows iOS Local
+ Cloud
Superwhisper is a voice dictation app for macOS, Windows and iOS offering both local (offline) and cloud AI transcription. Priced at $8.49/month or $249.99 lifetime. This independent review covers WER/CER accuracy across 6 test recordings, a privacy analysis, and a UX verdict.
Superwhisper Verdict
Powerful local engine buried under a broken cloud flagship
Superwhisper version 1.4.0 scores 6.3/10 overall in Voice-list independent testing (tested 2026-05-30). Best local model (Whisper Standard) achieves 2.0% aggregate WER across 6 recordings. Best cloud model (Ultra) achieves 2.4% aggregate WER.
Works well for
- Whisper Standard (hidden): best-in-class local accuracy at 1.8% WER
- Genuine offline mode — no audio leaves device in local configuration
- Lifetime license option at $249.99 — no subscription required
Watch out for
- S1-Voice (cloud flagship default) shows 15-37% WER across test recordings
- Trial is 15 minutes total — blocks even offline models after limit
- Cloud mode sends app name, clipboard, and focused text to Modal.com beyond audio
Best for
- Power users willing to dig past default settings and switch to Whisper Standard
Not for
- Anyone who installs and expects the default model to work well
Superwhisper Accuracy & Speed
| Model | Accuracy | Speed | ||
|---|---|---|---|---|
| English | Local | Parakeet Default 1.1 GB CPU Tested on CPU Ryzen AI 9 HX · 32 GB RAM
NVIDIA Parakeet TDT 0.6B — default local model in Superwhisper 1.4. Fast, good on clean speech, but has no ITN: numbers and dates come out as spoken words. Default local model — users see this first | 82.3%
Word accuracy
The share of words the model got right (100% − word error rate). 100% = every word correct.
17.7% WER Word Error Rate
What % of words the model got wrong. 0% = every word correct.
13.7% CER Character Error Rate
Same as WER but measured letter-by-letter. Usually lower than WER.
25% PER Punctuation Error Rate
How accurately the model placed commas, periods, and other punctuation.
4 / 10 |
~3s
2–5s range
Post-stop latency
Seconds from pressing Stop to the final text appearing in your active app. Average across all test recordings.
8 / 10 |
| Whisper Standard Best accuracy hidden in UI
500 MB CPU Tested on CPU Ryzen AI 9 HX · 32 GB RAM
OpenAI Whisper large-v2 running locally. Best accuracy of all tested models but hidden from the main model picker — requires Library search to find. Hidden — Settings > Library > search "Whisper Standard" | 98.0%
Word accuracy
The share of words the model got right (100% − word error rate). 100% = every word correct.
2.0% WER Word Error Rate
What % of words the model got wrong. 0% = every word correct.
0.5% CER Character Error Rate
Same as WER but measured letter-by-letter. Usually lower than WER.
35% PER Punctuation Error Rate
How accurately the model placed commas, periods, and other punctuation.
9 / 10 |
~8s
6–12s range
Post-stop latency
Seconds from pressing Stop to the final text appearing in your active app. Average across all test recordings.
5 / 10 | ||
| Cloud | S1-Voice Default Superwhisper's proprietary cloud model, presented as the headline AI feature. Applies aggressive rewriting that causes large content losses on some recordings. Default cloud model — users land here without changing settings | 74.2%
Word accuracy
The share of words the model got right (100% − word error rate). 100% = every word correct.
25.8% WER Word Error Rate
What % of words the model got wrong. 0% = every word correct.
21.8% CER Character Error Rate
Same as WER but measured letter-by-letter. Usually lower than WER.
67% PER Punctuation Error Rate
How accurately the model placed commas, periods, and other punctuation.
2 / 10 |
~2s
1–4s range
Post-stop latency
Seconds from pressing Stop to the final text appearing in your active app. Average across all test recordings.
9 / 10 | |
| Ultra Best Cloud Cloud Whisper-class model offered as "Ultra" tier. More conservative post-processing than S1-Voice — accurate and stable across all recording types. Better than S1-Voice on every recording — less prominently surfaced in UI | 97.6%
Word accuracy
The share of words the model got right (100% − word error rate). 100% = every word correct.
2.4% WER Word Error Rate
What % of words the model got wrong. 0% = every word correct.
1.0% CER Character Error Rate
Same as WER but measured letter-by-letter. Usually lower than WER.
40% PER Punctuation Error Rate
How accurately the model placed commas, periods, and other punctuation.
8 / 10 |
~3s
2–5s range
Post-stop latency
Seconds from pressing Stop to the final text appearing in your active app. Average across all test recordings.
8 / 10 | ||
| No models match — turn a filter back on. | ||||
Superwhisper for Coding & IT Best: Whisper Standard Ultra
Ultra Cloud
Coding 92.6% 17 err / 230w
Conference 99.5% 1 err / 220w
Coding
- snake_case identifiers mostly preserved
- CLI flags correct
- No hallucinations
- "Tauri" → "Tory"
- "tokio::runtime" → "Toko runtime"
alternative wrong extra missing
okay ,so .the deploy pipeline is broken again .um ,let me walk through what we have .the repo is a rust workspace with a atari front-end and an axm back-end .cargo.tomel at the root .two members ,desktop and server .the postgres migration that landed last tuesday added a new column to the users table called um ,last underscore scene underscore at type timestamp with timezone default now diesel picked it up fine but sqlx in the server crate started failing the macro expansion at compile time because uh ,the offline query metadata json was stale run cargo sqlx prepare workspace before pushing ok ,now the kubernetes side the helm chart for the staging cluster bumped the image tag to ,um ,v1.4.0 alpha 3 ,but the deployment yaml still references v1.3.11 .that mismatch is what is throwing the image pullback off .update values.yaml under environments slash staging .the github actions workflow is using ubuntu 2204 runners ,but rust-toolchain.toml is pinned to 1.83 ,and the runner stable is 1.79 ,so either bump the runner image or pin rustup install to 1.83 explicitly .for the cloudflare worker ,wrangler.toml has the wrong account underscore id after we migrated the team ,and miniflare is failing the d1 binding lookup .run wrangler login again .finally ,explain analyze on the sessions token lookup .p95 should stay under 15 milliseconds .if it regresses ,the missing index on the ,um ,jsonb metadata column is the prime suspect .
Conference
- Best on accented English speaker — 0.45% WER
- "Kubernetes", "PostgreSQL", "microservices" exact
- Zero dropped sentences
- "load balancer" → "load balance" (once)
alternative wrong extra missing
so ,the question i want to address today is whether on-device speech recognition is finally good enough to replace the cloud .the short answer is yes ,for english at least ,and the longer answer is the interesting part .when openai released whisper in ,um ,2022 ,the smallest models were already competitive with cloud apis for clean audio ,but they were slow .three years later ,two things have changed .first ,the c++ port from georgi gurganov ,called whisper.cpp ,runs the same model significantly faster than the python reference .and ,ah ,it runs on a raspberry pi .second ,nvidia released parakeet tdt in 2025 ,which is between 5 and 10 times faster than whisper on cpu and has similar accuracy for english .so now ,the question is not ,can we run speech recognition locally ?but ,um ,should we and what do we give up ?the honest answer is that you give up the ability to update the model continuously ,which cloud providers can do every week ,and you give up some accuracy on the long tail of less common languages .what you gain is ,you know ,privacy by architecture rather than by policy ,lower latency for short phrases on modern hardware and zero per minute cost at scale .for most desktop dictation use cases in 2026 ,that trade-off is the right one .
Whisper Standard Local
Coding 92.7% 17 err / 233w
Conference 99.1% 2 err / 220w
Coding
- snake_case identifiers intact
- CLI flags like "--release" correct
- No hallucinations
- "Tauri" → "Tarry"
- "axum" → "axm"
alternative wrong extra missing
ok ,so the deploy pipeline is broken again .um ,let me walk through what we have .the repo is a rust workspace with a tarry front end and an axum back end .cargo.toml at the root .two members ,desktop and server .the postgres migration that landed last tuesday added a new column to the user's table called ,um ,last_scene_ at type_timestamp with timezone default now .diesel picked it up fine ,but sqlx and the server crate started failing the macro expansion at compile time ,because ,uh ,the offline query metadata ,json was stale .run cargo sqlx ,prepare workspace before pushing .ok ,now the kubernetes side .the helm chart for the staging cluster bumped the image tag to ,um ,v1.4.0 alpha 3 ,but the deployment yaml still references v1.3.11 .that mismatch is what is throwing the image pullback off .update values .yaml under environments / staging .the github actions workflow is using ubuntu 2204 runners ,but rusttoolchain.toml is pinned to 1.83 ,and ,uh ,the runner stable is 1.79 ,so either bump the runner image or pin rustup install to 1.83 explicitly .for the cloudflare worker ,wrangler.toml has the wrong account underscore id after we migrated the team ,and miniflare is failing the d1 binding lookup .run wrangler login again .finally ,explain analyze on the session's token lookup .p95 should stay under 15 milliseconds .if it regresses ,the missing index on the um ,jsonb metadata column is the prime suspect .
Conference
- Strong on accented English
- "PostgreSQL" and "Kubernetes" exact
- Only 1 minor substitution across 89s
- "schema" → "schema" (case flip once)
alternative wrong extra missing
so ,the question i want to address today is whether on-device speech recognition is finally good enough to replace the cloud .the short answer is yes ,for english at least ,and the longer answer is the interesting part .when openai released whisper in ,um ,2022 ,the smallest models were already competitive with cloud apis for clean audio ,but they were slow .three years later ,two things have changed .first ,the c++ ,port from georgi gurganov ,called whisper.cpp ,runs the same model significantly faster than the python reference ,and ,uh ,it runs on a raspberry pi .second ,nvidia released parakeet ttt in 2025 ,which is between 5 and 10 times faster than whisper on cpu and has similar accuracy for english .so now ,the question is not ,can we run speech recognition locally ,but ,um ,should we ,and what do we give up ?the honest answer is that you give up the ability to update the model continuously ,which cloud providers can do every week ,and you give up some accuracy on the long tail of less common languages .what you gain is ,you know ,privacy by architecture ,rather than by policy ,lower latency for short phrases on modern hardware ,and zero per minute cost at scale .for most desktop dictation use cases in 2026 ,that trade-off is the right one ,
Parakeet Local
Coding 83.3% 39 err / 234w
Conference 98.2% 4 err / 222w
Coding
- Handles prose segments cleanly
- No hallucinations
- "cargo.toml" → "Cargo .toml"
- "tokio::spawn" → "tokio spawn"
- "impl Trait" → "imp Trait"
alternative wrong extra missing
okay ,so the deploy pipeline is broken again .um ,let me walk through what we have .the repo is a rust workspace ,with a atari front end and uh an axum backend .cargo.toml at the cargo .two members ,desktop and server .the postgres migration that landed last tuesday added a new column to the user's table called um last underscore scene ,underscore at type timestamp with time zone ,default now .diesel picked it up fine ,but sqlx in the server crate started failing the macro expansion at compile time ,because uh the offline query metadata ,json was stale .run cargo sqlx prepare workspace before pushing .okay ,now the kubernetes side .the helm chart for the staging cluster bumped the image tag to um v1 .4 .0 alpha 3 ,but the deployment yaml still references v1 .3 .11 .that mismatch is what is throwing the image pullback off .update values .yaml under environments slash staging .the github actions workflow is using ubuntu 2204 runners ,but rust toolchain .toml is pinned to 1 .83 ,and uh the runner stable is 1 .79 ,so either bump the runner image ,or pin rustup install to 1 .83 explicitly .for the cloudflare worker ,wrangler .toml has the wrong account underscore id after we migrated the team ,and miniflare is failing the d1 binding lookup ,run wrangler login again .finally ,explain analyze on the sessions token lookup ,p95 should stay under 15 milliseconds .if it regresses ,the missing index on the um jsonb metadata column is the prime suspect .
Conference
- Clean transcription of accented English speaker
- Technical terms like "API" and "SDK" correct
- "Kubernetes" → "Cubernetes"
alternative wrong extra missing
so ,the question i want to address today is whether on device speech recognition is finally good enough to replace the cloud .the short answer is yes ,for english at least ,and the longer answer is the interesting part .when openai released whisper in um 2022 ,the smallest models were already competitive with cloud apis for clean audio ,but they were slow .three years later ,two things have changed .first ,the c port from georgi gurganov called whisper .cpp runs the same model significantly faster than the python reference .and uh it runs on a raspberry pi .second ,nvidia released parakeet tdt in 2025 ,which is between five and ten times faster than whisper on cpu and has similar accuracy for english .so now ,the question is not can we run speech recognition locally ,but um should we ,and what do we give up ?the honest answer is that you give up the ability to update the model continuously ,which cloud providers can do every week ,and you give up some accuracy on the long tail of less common languages .what you gain is ,you know ,privacy by architecture rather than by policy ,lower latency for short phrases on modern hardware ,and zero per minute cost at scale .for most desktop dictation use cases in 2026 that trade off is the right one .
S1-Voice Cloud
Coding 64.3% 81 err / 230w
Conference 75.5% 54 err / 220w
Coding
- Faster response than local models (~2s)
- "impl AsyncRead" → "RUMP_INSTAL"
- Dropped entire code block (lines 14–17)
- "CloudFace" hallucinated (not in source)
alternative wrong extra missing
okay so .the deploy pipeline is broken again .um let me walk through what we have :the repo is a rust workspace with a atari front end and an axm backend .cargo dot tomel at the root ;two members desktop and server .the postgres migration that landed last tuesday added a new column to the users table called um last_seen underscore at type timestamp with time zone .default now .diesel picked it up fine but sqlx in the server crate started failing the macro expansion at compile time because ,uh .the offline query metadata json was stale .run cargo sqlx prepare workspace before pushing .the kubernetes github actions workflow is using ubuntu 220.4 runners but rusttoolchain.toml is pinned to um v1.4.0 alpha 1.83 and the runner stable is 1.79 so either bump the runner image or pin rustup install to 1 point 83 explicitly .for the cloudface ,the rump_instal is a bit of a runaway .the rupp_instr is a little bit more than one day .the runner stable is has been pinned to the runner image .the rustup is pinned down to 1 0.83 explicitly .as a flare worker ,wrangler.toml has the wrong account_id after we migrated the team ,and miniflare is failing the d1 binding lookup .run wranglar login again .finally ,explain analyze on the sessions token lookup :p95 should stay under 15 milliseconds .if it regresses ,the missing index on the ,um ,jsonb metadata column is the prime suspect .
Conference
- Fast turnaround even on longer clip
- "distributed systems" → "destructive systems"
- Dropped 3 full sentences mid-recording
- Non-deterministic: 37% WER run 1 vs 16% run 2 — same audio
alternative wrong extra missing
so .the question i want to address today is whether on-device speech recognition is finally good enough to replace the cloud ?the short answer is yes - for english at least .and the longer answer is the interesting part .when openai released whisper in .um 2022 ,the smallest models were already competitive with cloud apis for clean audio but they were slow .three years later two things have changed :first the c++ port from georgi gerganov called whisper.cpp runs the same model significantly faster than the python reference and it runs on a raspberry pi 2 nvidia released parakeet tdt so now ,which is between five and ten times faster than whisper on cpu and has similar accuracy for english so now the question is not can we run speech recognition locally ?but um .should we ?and what do we give up ?the honest answer is that you give up the ability to update the model continuously which cloud providers can do every week .and you give give-up some accuracy on the long tail of less common languages .what you gain is ,you know ,privacy by architecture rather than by policy ,lower latency for short phrases on modern hardware and zero per minute cost at scale .for most desktop dictation use cases in 2026 that trade off is the right one
Superwhisper for Everyday & Long-form Best: Whisper Standard Ultra
Ultra Cloud
Casual 100.0% 0 err / 190w
Long-form 99.8% 1 err / 543w
Casual
- Perfect: zero errors on casual speech
- Natural disfluency handling
alternative wrong extra missing
hey ,so i wanted to follow up on what we discussed yesterday about the onboarding flow .i think there are three separate problems and we have been treating them as one .first ,the email verification step has a 48% drop-off rate ,which means almost half of people who sign up never actually finish .that alone is worth looking at before we touch anything else .second the time zone selection screen is buried inside settings and people who skip it end up with notifications at three in the morning which is you know the worst possible experience we need to bring that into the main flow third the welcome email goes out two hours after sign up which feels kind of random it should go out either immediately while the user is still excited ,or i mean strategically the next morning ,but not at this awkward in-between moment .so my proposal is ,in order of impact ,fix the email verification first ,then move the time zone picker ,then retime the welcome email .what do you think ?let me know if you want to grab 15 minutes tomorrow to talk through it .
Long-form
- Best long-form result of any engine tested — near-perfect over 4 minutes
- No drift, zero hallucinations
- All numbers, currency and percentages formatted correctly
- One extra "and" — the only error in the whole recording
alternative wrong extra missing
okay ,so i want to walk through what we learned this quarter about organic search versus paid acquisition .because i think the numbers genuinely changed my mind .and i want the whole team aligned before we set the budget for next year .so ,quick context .for the last three years ,we have been spending roughly 70% of our marketing budget on paid channels .google ads ,a bit of meta ,some linkedin for the enterprise segment .the remaining 30% went into content and seo .and the assumption ,honestly ,was that paid is the reliable engine and seo is a slow ,nice-to-have thing on the side .turns out that framing was wrong .so let me give you the actual numbers .on paid ,our blended cost per acquisition climbed from $41 in january to $68 by september .that is a 66% increase in 9 months and nothing about our targeting changed .the auction just got more expensive .more competitors bidding on the same keywords plus the platform raising minimum bids .meanwhile ,our organic traffic went from about 12,000 sessions a month to 41,000 . and the cost per acquisition on that channel ,if you amortize the content investment ,was around $9 .9 versus 68 .that is not a small gap .now , the honest counter-argument is timing .paid converts today .you spend $1,000 on tuesday . you get leads on tuesday .seo is a delayed engine .the articles we published in february did not really start ranking until may or june . so there is a real cash flow difference .and if you are a startup that needs pipeline this month ,you cannot just turn off paid and wait two quarters for organic to compound .i get that .but here is the thing that surprised me .when we looked at lead quality ,not just volume ,the organic leads had a 31% higher trial-to-paid conversion rate .the theory is that someone who finds you by searching for a specific problem is further along in intent than someone who clicks an ad in their feed .they are actively looking .so not only is organic cheaper per lead ,the leads are actually better .so what are we doing differently next year ?three things .first ,we are flipping the ratio ,moving to roughly 50-50 between paid and organic over the next two quarters ,not all at once . because we still need the near-term pipeline .second ,we are doubling the content team from two writers to four . and we are focusing on what we call bottom of funnel comparison content ,because that is where the intent and the conversion rate are highest .and third ,we are going to treat paid as an accelerant for content that is already ranking instead of a standalone channel .so when an article hits page one organically ,we put paid behind it to compress the timeline .the goal by end of next year is to get our blended cost per acquisition back under $30 and to have organic driving more than half of all qualified pipeline .right now , it is at about 22% .that is a big gap to close ,but the trajectory over the last six months tells me it is achievable .anyway ,that is the short version .we can dig into the channel level breakdown in a separate session .
Whisper Standard Local
Casual 100.0% 2 err / 190w
Long-form 98.9% 8 err / 549w
Casual
- Perfect: zero word errors
- Disfluencies handled cleanly
alternative wrong extra missing
hey ,so i wanted to follow up on what we discussed yesterday about the onboarding flow .i think there are three separate problems and we have been treating them as one .first ,the email verification step has a 48% drop-off rate ,which means almost half of people who sign up never actually finish .that alone is worth looking at before we touch anything else .second ,the time zone selection screen is buried inside settings and people who skip it end up with notifications at 3 in the morning ,which is ,you know ,the worst possible experience .we need to bring that into the main flow .third ,the welcome email goes out two hours after sign up ,which feels kind of random .it should go out either immediately while the user is still excited or ,or i mean ,strategically the next morning ,but not at this awkward in-between moment .so my proposal is ,in order of impact ,fix the email verification first ,then move the time zone picker ,then retime the welcome email .what do you think ?let me know if you want to grab 15 minutes tomorrow to talk through it .
Long-form
- No drift over 4 minutes — consistent quality throughout
- Zero hallucinations across full recording
- "LinkedIn" split into "linked in"
- "$9. 9 versus" collapsed into "$9.9 versus"
- "counterargument" spelled as two words
alternative wrong extra missing
okay ,so i want to walk through what we learned this quarter about organic search versus paid acquisition because i think the numbers genuinely changed my mind and i want the whole team aligned before we set the budget for next year .so ,quick context .for the last three years ,we have been spending roughly 70% of our marketing budget on paid channels .google ads ,a bit of meta ,some linked in for the enterprise segment .the remaining 30% went into content and seo and the assumption ,honestly ,was that paid is the reliable engine and seo is a slow ,nice to have thing on the side .turns out that framing was wrong .so ,let me give you the actual numbers .on paid ,our blended cost per acquisition climbed from $41 in january to $68 by september .that is a 66% increase in nine months and nothing about our targeting changed .the auction just got more expensive ,more competitors bidding on the same keywords ,plus the platform raising minimum bids .meanwhile ,our organic traffic went from about 12,000 sessions a month to 41,000 . and the cost per acquisition on that channel ,if you amortized the content investment ,was around $9 $9.9 versus 68 .that is not a small gap .now , the honest counter argument is timing .paid converts today .you spend $1,000 on tuesday ,you get leads on tuesday .seo is a delayed engine .the articles we published in february did not really start ranking until may or june . so , there is a real cash flow difference .and if you are a startup that needs pipeline this month ,you cannot just turn off paid and wait two quarters for organic to compound .i get that .but here is the thing that surprised me .when we looked at lead quality ,not just volume ,the organic leads had a 31% higher trial to paid conversion rate .the theory is that someone who finds you by searching for a specific problem is further along in intent than someone who clicks an ad in their feed .they are actively looking .so , not only is organic cheaper per lead ,the leads are actually better .so , what are we doing differently next year ?three things .first ,we are flipping the ratio ,moving to roughly 50-50 between paid and organic over the next two quarters ,not all at once . because ,ah , we still need the near term pipeline .second ,we are doubling the content team from two writers to four . and we are focusing on what we call bottom of funnel comparison content . because that is where the intent and the conversion rate are highest .and third ,we are going to treat paid as an accelerant for content that is already ranking instead of a standalone channel .so , when an article hits page one organically ,we put paid behind it to compress the timeline .the goal by end of next year is to get our blended cost per acquisition back under $30 and to have organic driving more than half of all qualified pipeline .right now , it is at about 22% .that is a big gap to close . but the trajectory over the last six months tells me it is achievable .anyway ,that is the short version .we can dig into the channel level breakdown in a separate session .
Parakeet Local
Casual 100.0% 0 err / 197w
Long-form 98.6% 8 err / 559w
Casual
- Perfect: zero word errors on casual speech
- Disfluencies (um, uh) preserved naturally
alternative wrong extra missing
hey ,so i wanted to follow up on what we discussed yesterday about the onboarding flow .i think there are um three separate problems ,and we have been treating them as one .first ,the email verification step has a 48 % drop off rate ,which means almost half of people who sign up never actually finish .that alone is worth looking at before we touch anything else .second ,the time zone selection screen is uh buried inside settings ,and people who skip it end up with notifications at three in the morning ,which is ,you know ,the worst possible experience .we need to bring that into the main flow .third ,the welcome email goes out two hours after sign up ,which feels um kind of random .it should go out either immediately while the user is still excited ,or i mean strategically the next morning ,but not at this awkward in between moment .so my proposal is in order of impact ,fix the email verification first ,then move the time zone picker ,then re time the welcome email .what do you think ?let me know if you want to grab 15 minutes tomorrow to talk through it .
Long-form
- No quality drift over 4 minutes — consistent throughout
- Numbers and currency mostly formatted correctly
- "numbers genuinely" mangled into "numbinely"
- "50/50" misheard as "550"
- "accelerant" → "acceleration"; dropped "trial" and "through"
alternative wrong extra missing
okay ,so i want to walk through what we learned this quarter about organic search versus paid acquisition because i think the numbers numbinely changed my mind ,and i want the whole team aligned before we set the budget for next year .so quick context .for the last three years ,we have been spending roughly 70 % of our marketing budget on paid channels .google ads ,a bit of meta ,some linkedin for the enterprise segment .the remaining 30 % went into content and seo .and the assumption ,honestly ,was that paid is the reliable engine ,and seo is a slow nice to have thing on the side .turns out that framing was wrong .so let me give you the actual numbers .on paid ,our blended cost per acquisition climbed from 41 in january to 68 by september .that is a 66 % increase in nine months ,and nothing about our targeting changed .the auction just got more expensive .more competitors bidding on the same keywords ,plus the platform raising minimum bids .meanwhile ,our organic traffic went from about 12 , 000 sessions a month to 41 , 000 . and the cost per acquisition on that channel ,if you amortize the content investment was around $9 .9 versus 68 .that is not a small gap .now , uh the honest counterargument is timing .paid converts today .you spend a thousand dollars on tuesday ,you get leads on tuesday .seo is a delayed engine .the articles we published in february did not really start ranking until may or june . so there is a real cash flow difference .and if you are a startup that needs pipeline this month ,you cannot just turn off paid and wait two quarters for organic to compound .i get that .but here is the thing that surprised me .when we looked at lead quality ,not just volume ,the organic leads had a 31 % higher trial to paid conversion rate .the theory is that someone who finds you by searching for a specific problem is further along in intent than someone who clicks an ad in their feed .they are actively looking .so not only is organic cheaper per lead ,the leads are actually better .so what are we doing differently next year ?three things .first ,we are flipping the ratio ,moving to roughly 550 between paid and organic over the next two quarters ,not all at once . because we still need the near term pipeline .second ,we are doubling the content team from two writers to four . and we are focusing on what we call bottom of funnel comparison content ,because that is where the intent and the conversion rate are highest .and third ,we are going to treat paid as an acceleration for content that is already ranking instead of a standalone channel .so when an article hits page one organically ,we put paid behind it to compress the timeline .the goal by end of next year is to get our blended cost per acquisition back under 30 dollars ,and to have organic driving more than half of all qualified pipeline .right now , it is at about 22 % .that is a big gap to close . but the trajectory over the last six months tells me it is achievable .anyway ,that is the short version .we can dig into the channel level breakdown in a separate session .
S1-Voice Cloud
Casual 63.0% 71 err / 192w
Long-form 81.2% 105 err / 549w
Casual
- Lost 2 of 3 sections — only opening paragraph survived
- Heavy rewriting distorts meaning of what remains
alternative wrong extra missing
hey !so i wanted to follow up on what we discussed yesterday about the onboarding flow .i think there are three separate problems and we have been treating them as one :first ,the email verification step has a 48% drop-off rate which means almost half of people who sign up never actually finish .that alone is worth looking at before we touch anything else .and the time zone selection screen is buried inside settings .and people who skip it end up with notifications at 3:00 in the morning which is you know the worst possible experience .we need to bring that into the main flow third the welcome email goes out two hours after sign up which feels um kind of random .it should go out either immediately while the user is still excited or i mean strategically the next morning but not at this awkward in-between moment so my proposal is in order of impact to fix the email verification first then move the time zone picker ,then re-time the welcome email what do you think let me know if you want to grab 15 minutes tomorrow to talk through it .
Long-form
- Completes the full recording without timeout
- Catastrophic drift mid-recording — hallucinated a whole passage ("a range of results... the last one was a long time ago")
- Dropped the entire "let me give you the actual numbers" section
- Worst long-form result of any engine tested — 18.9% WER
alternative wrong extra missing
ok so i want to walk through what we learned this quarter about organic search versus paid acquisition .because i think the numbers genuinely changed my mind and i want the whole team aligned before we set the budget for next year .so quick context :for the last three years we have been spending roughly 70 percent of our marketing budget on paid channels .google ads a bit of meta ,some linkedin for the enterprise segment the remaining 30% went into content and seo .and the assumption honestly was that paid is the reliable engine and seo is a slow nice to have thing on the side .turns out that framing was wrong let me give you the actual numbers .on paid our blended cost per acquisition climbed from 41 in january to $68 by september .that is a 66% increase in 9 months .and nothing about our targeting changed .the auction just got more expensive :more competitors bidding on the same keywords plus the platform raising minimum bids .meanwhile our organic traffic went from about 12,000 sessions a month to 41,000 . and the cost per acquisition on that channel ? if you amortize the content investment was around nine dollars 9 versus $68 that is not a small gap now the honest counter argument is timing .paid converts today : you spend a thousand dollars on tuesday ; you get leads on tuesday .seo is a delayed engine .the articles we published in february did not really start ranking until may or june . so there's a range of results that are going to be the same for you . the last one was a long time ago . the first two weeks ago ,the last two weeks of the year was a week ago . and then the last one came out with the first two months ago . there's a real cash flow difference .and if you are a startup that needs pipeline this month ,you cannot just turn off paid and wait two quarters for organic to compound .i get that ! but here is the thing that surprised me : when we looked at lead quality not just volume ,the organic leads had a 31% higher trial-to-paid conversion rate .the theory is that someone who finds you by searching for a specific problem is further along in intent than someone who clicks an ad in their feed .they are actively looking , so not only is organic cheaper per lead the leads are actually better .so what are we doing differently next year ?three things : first we are flipping the ratio moving to roughly 50/50 between paid and organic over the next two quarters not all at once because ah still need the near-term pipeline ! second we are doubling the content team from two writers to four . and we are focusing on what we call bottom of funnel comparison content because that is where the intent and the conversion rate are highest .and third we are going to treat paid as an accelerant for content that is already ranking instead of a standalone channel so when an article hits page one or we put paid behind it to compress the timeline .the goal by end of next year is to get our blended cost per acquisition back under $30 and to have organic driving more than half of all qualified pipeline .right now it is at about 22% .that is a big gap to close ,but the trajectory over the last six months tells me it is achievable .anyway that is the short version .we can dig into the channel level breakdown in a separate session .
Superwhisper for Numbers & Structured Data Best: Whisper Standard Ultra
Whisper Standard Local
Numbers/ITN 97.5% 3 err / 40w
Numbers/ITN
- Dates and currency nearly exact
- "$12,400.75" and phone number correct
- "March 15th, 2026" → "March 15, 2026" (minor format)
alternative wrong extra missing
okay ,the meeting is on march 15 ,2026 at 3.30 p.m .pacific time .the proposed budget is $12,400.75 which is about 1.6% under the original estimate .phone number for the conference bridge is 1-800-555-0123 extension 479 .order id is abc-1 23456 .
Ultra Cloud
Numbers/ITN 94.9% 2 err / 39w
Numbers/ITN
- Phone number and date format correct
- "$12,400.75" → "$12400.75" (comma dropped)
- "Order ID" label partially dropped
alternative wrong extra missing
okay ,the meeting is on march 15th ,2026 at 3:30 3.30pm pst .the proposed budget is $12,400.75 which is about 1.6% under the original estimate .phone number for the conference bridge is 1-800-555-0123 extension 479 .order id is abc-123456 .
S1-Voice Cloud
Numbers/ITN 92.7% 3 err / 41w
Numbers/ITN
- "$12,400.75" exact
- Phone number format correct
- "March 15th, 2026" → "March 15, 2026"
- "ABC-123456" → "ABC 123456" (hyphen dropped)
alternative wrong extra missing
okay .the meeting is on march 15th ,2026 at 3:30pm pacific time .the proposed budget is $12,400.75 which is about 1.6% under the original estimate .phone number for the conference bridge is 1800 555 0123 extension 479 .order id is abc-123456 .
Parakeet Local
Numbers/ITN 14.0% 37 err / 43w
Numbers/ITN
- No ITN — numbers output as spoken words throughout
- "$12,400.75" → "twelve thousand four hundred dollars and seventy five cents"
- Phone number and order ID completely garbled
alternative wrong extra missing
okay .the meeting is on march fifteenth ,twenty six ,at three thirty p .m .pacific time .the proposed budget is twelve thousand four hundred dollars and seventy five cents ,which is about one point six percent under the original estimate .phone number for the conference bridge is one eight hundred five zero one two three extension four seven nine .order id is abc one two three four five six .
Superwhisper: Noise Resistance Best: Parakeet Ultra
Parakeet Local
Noisy Cafe 100.0% 0 err / 195w
Noisy Cafe
- Noise has zero effect — identical output to clean version
- Café background at SNR 5 dB not detected
alternative wrong extra missing
hey ,so i wanted to follow up on what we discussed yesterday about the onboarding flow .i think there are um three separate problems and we have been treating them as one .first ,the email verification step has a 48 % drop off rate ,which means almost half of people who sign up never actually finish .that alone is worth looking at before we touch anything else .second ,the time zone selection screen is uh buried inside settings ,and people who skip it end up with notifications at three in the morning ,which is you know the worst possible experience .we need to bring that into the main flow .third ,the welcome email goes out two hours after sign up ,which feels um kind of random .it should go out either immediately while the user is still excited ,or i mean strategically the next morning ,but not at this awkward in-between moment .so my proposal is in order of impact ,fix the email verification first ,then move the time zone picker ,then re-time the welcome email .what do you think ?let me know if you want to grab 15 minutes tomorrow to talk through it .
Whisper Standard Local
Noisy Cafe 100.0% 0 err / 190w
Noisy Cafe
- Noise has zero effect — identical output to clean version
alternative wrong extra missing
hey ,so i wanted to follow up on what we discussed yesterday about the onboarding flow .i think there are three separate problems and we have been treating them as one .first ,the email verification step has a 48% drop-off rate ,which means almost half of people who sign up never actually finish .that alone is worth looking at before we touch anything else .second ,the time zone selection screen is buried inside settings and people who skip it end up with notifications at 3 in the morning ,which is ,you know ,the worst possible experience .we need to bring that into the main flow .third ,the welcome email goes out two hours after sign up ,which feels kind of random .it should go out either immediately while the user is still excited ,or i mean strategically the next morning ,but not at this awkward in-between moment .so my proposal is in order of impact ,fix the email verification first ,then move the time zone picker ,then retime the welcome email .what do you think ?let me know if you want to grab 15 minutes tomorrow to talk through it .
Ultra Cloud
Noisy Cafe 99.0% 4 err / 190w
Noisy Cafe
- Near-perfect under café noise — only 1 minor substitution
- "in-between" split once
alternative wrong extra missing
hey ,so i wanted to follow up on what we discussed yesterday about the onboarding flow .i think there are three separate problems and we have been treating them as one .first ,the email verification step has a 48% drop-off rate ,which means almost half of people who sign up never actually finish .that alone is worth looking at before we touch anything else .second ,the time zone selection screen is buried inside settings and people who skip it end up with notifications at 3 in the morning which is you know the worst possible experience .we need to bring that into the main flow .third ,the welcome email goes out two hours after sign up which feels kind of random .it should go out either immediately while the user is still excited or ,or i mean ,strategically the next morning .but not at this awkward in-between moment .so my proposal is ,in order of impact ,fix the email verification first ,then move the time zone picker ,then re-time the welcome email .what do you think ?let me know if you want to grab 15 minutes tomorrow to talk through it .thank you .
S1-Voice Cloud
Noisy Cafe 95.2% 7 err / 188w
Noisy Cafe
- Handles café noise better than casual clean — rewriting helps here
- "in-between" → "in between"
- Some filler words not stripped
alternative wrong extra missing
hey !so i wanted to follow up on what we discussed yesterday about the onboarding flow .i think there are three separate problems and we have been treating them as one .first ,the email verification step has a 48% drop-off rate which means almost half of people who sign up never actually finish that alone is worth looking at before we touch anything else .second the time zone selection screen is buried inside settings .and people who skip it end up with notifications at 3:00 in the morning which is you know the worst possible experience .we need to bring that into the main flow .third ,the welcome email goes out two hours after sign-up which feels kind of random .it should go out either immediately while the user is still excited or i mean strategically i'm not going to go early the next morning but not at this awkward in-between moment .so my proposal is :in order of impact fix the email verification first then move the timezone picker then retime the welcome email .what do you think ?let me know if you want to grab 15 minutes tomorrow to talk through it .
Tested on Windows 11 26H2 · AMD Ryzen AI 9 HX 370 · 32 GB RAM · NVIDIA RTX 5070 Laptop 8 GB
Superwhisper UX & Integration
Getting started & flow
Onboarding flow
Reached first successful dictation in about a minute — nothing superfluous.
Hotkey customization
Default shortcut is comfortable and remappable, no system conflicts — but the push-to-talk option does not actually work.
Error messages
Shows a center-screen message when the trial runs out, but there is no fallback — and settings navigation is scattered across sections.
Recording experience
Recording overlay UX
Clear recording pill / overlay — recording state is obvious.
Stop / cancel UX
Easy to cancel a bad dictation; cancel hotkey included.
Text insertion reliability
Pastes reliably into every app tested.
Auto-insert vs clipboard
Auto-inserts the text and can restore your previous clipboard afterwards.
Managing your work
Recording history
Browsable history with search; you can open a recording to see its mode, duration and even the prompt used. No export.
Mode / model switching
Fast switching by hotkey and from the pill UI.
Idle resource use
~160 MB RAM · 0.3% CPU at rest (cloud).
Superwhisper Features
Text processing
AI post-processing
Cloud AI modes rewrite text — many models, BYOK for several providers. But S1-Voice over-rewrites and drops content.
Custom vocabulary / dictionary
Per-word replacements applied at transcription.
Text snippets / expansion
Bundled into the custom-dictionary feature, not a separate snippets UI — and also doable via LLM post-processing instructions.
Output & extras
File transcription
Hidden behind the tray icon. Broken on LLM modes (returns a stale buffer); only works on the Voice mode, and the UX is so confusing it barely counts.
Music auto-mute
Pause, lower, or fully mute media while recording.
Translation mode
No built-in translation mode.
Ask / Q&A mode
No Ask / Q&A LLM mode.
Export (txt / srt / json)
No txt / srt / json export, and history cannot be bulk-exported.
Voice commands
Local recognition
Offline / local inference
Genuine offline mode — no audio leaves device in local configuration.
Multiple model options
Parakeet, Whisper Standard (hidden), S1-Voice, Ultra — but best local model is buried.
Superwhisper Privacy
Superwhisper keeps audio on-device when using local models. Cloud models upload audio to modal.com only after you press Stop.
Audio uploaded in cloud mode only
Endpoints: modal.com, api.superwhisper.com
Audio sent only after you press Stop
Nothing is uploaded until you confirm by pressing Stop. Cancel before then and the audio never leaves.
Account optional
You can use the app without an account, but some features ask you to sign in.
Sends more than audio
In cloud mode: active app name, focused element text, clipboard contents, computer name, locale, timezone
Opt out of training on your data
Your recordings are not used to train models.
Disable analytics & tracking
You can turn off product analytics and telemetry.
Turn off history storage
You can set the app to never store your transcription history.
From the privacy policy not scored
- Privacy policy guarantees data is never used to train AI models and is not retained on Superwhisper servers — all storage is local.
- States it collects no usage data and uses no cookies or tracking technologies.
- Note: the observed cloud mode still sends app context and clipboard to Modal.com — stronger than the policy implies, so cloud users should not assume "local-only".
Pricing
Trial: 15-minute free trial — then basic features stay free, no payment needed
Free Trial Basic features after the trial
- Voice to text in any app
- Meeting recording and transcription
- Unlimited use of small AI models
- 100+ languages, custom prompt control
- 15-minute trial limit (all models)
Subscription $8.49/mo Pro · $84.99/yr (2 months free) · 40% student discount · 30-day refund
- Unlimited use of cloud and local AI models
- Bring your own AI API keys
- Translate any language to English
- Transcribe audio and video files, priority support
Lifetime $249.99one-time
- All Pro features, one-time payment — no subscription
- Activates across multiple devices
Superwhisper on the free tier
Superwhisper has no real free tier — only a 15-minute trial, after which a paid plan is required. It is excluded from our Best free option ranking. How we judge free tiers →
Methodology
Accuracy scores use WER (Word Error Rate) computed against multi-reference ground truth
with {a|b} alternates for valid transcription variants (e.g. 48% and
forty-eight percent are both accepted). Audio delivered via virtual cable from
ElevenLabs TTS. Single test session on 2026-05-30.
Superwhisper benchmarks
How Superwhisper places against every app we tested, on identical audio.
Superwhisper screenshots
Actual screenshots from our test session — click any image to view full size.
Superwhisper head-to-head
See Superwhisper compared point by point against the alternatives.