Introducing the S1 family of models
Introducing the S1 family of models
Nico DiPlacidoAug 19, 2026 · 7 min read
The S1 family of models has arrived.<br>At Superwhisper we believe that state-of-the-art, voice-to-text tools can be built to be faster, and more accurate without sacrificing your privacy by collecting or training on your data.<br>Today we’re excited to launch our proprietary Superwhisper models: S1-Voice, S1-Language, and S1-mini.<br>The S1 models have been trained and fine-tuned in-house at Superwhisper using bespoke synthetic, and publicly-available datasets that capture the hurdles of our customers: noisy office environments, whispered speech, and highly technical medical, financial, and mixed language dictations.<br>Each model is tuned from our customers feedback: S1-Voice turns speech into polished text in any application, S1-Language providing promptable inference to manipulate or clean up your dictations, and S1-mini offering tone-controlled, more-than-smart-enough clean up and reformatting achieved completely locally on your laptop or phone.<br>S1-mini: on-device cleanup, open weights<br>S1-mini is a small 484 MB 0.6B language model that uses local inference to output highly accurate, and tightly contextually aware formatted text with zero network requests.<br>We’re excited to release S1-mini with open weights now available on Hugging Face.<br>By design, S1-mini is highly focused to achieve a handful of common tasks extremely well. This allows us to have an efficient and lightweight model running locally on consumer devices.<br>Controlling your Tone preferences<br>After selecting S1-mini as your language model, you’ll have access to the new tone control slider with 5 preferences:<br>casual (all lowercase, minimal punctuation)<br>semi-casual<br>balanced<br>semi-formal<br>formal (contractions expanded, full punctuation)<br>Automatic formatting, text replacements, and mistake correction<br>S1-mini automatically detects lists content and transforms them into ordered or unordered lists for better visual clarity and readability. In mail mode, the model automatically contextually formats the transcript as an email with your greeting line, sign-off, and proper spacing:<br>Raw transcript:<br>hi Ben thanks for your patience while we looked into this we found three issues with the last shipment the packaging was damaged the wrong size was sent and the invoice amount was incorrect we already proc- we already processed a replacement and a refund for the difference let us know if you have any other questions thanks the support team
Superwhisper S1-mini returns:<br>Hi Ben,
Thanks for your patience while we looked into this. We found three issues with the last shipment:
• The packaging was damaged<br>• The wrong size was sent<br>• The invoice amount was incorrect
We already processed a replacement and a refund for the difference. Let us know if you have any other questions.
Thanks,<br>The Support Team
S1-mini also handles filler words, stutters, and corrections made while dictating by smartly backtracking and capturing what you intended to communicate. For instance, correcting yourself mid-dictation:<br>Raw transcript:<br>the meeting is on Tuesday I mean Thursday
Superwhisper S1-mini returns:<br>the meeting is on Thursday
It also knows what to leave alone: "three or four days" is a timeframe of a trip, and "tea or coffee" is a question rather than a mistake.<br>The model correctly and consistently renders numbers, dates, currency, percentages, phone numbers with country codes, spoken email addresses and URLs to make transcripts more readable. Say “support at superwhisper dot com" and you get support@superwhisper.com in return.<br>What it will not do<br>S1-mini is ruthlessly obedient. It will never add content you did not say, correct facts, soften profanity, flag what you are talking about, or rewrite your dialect. Its sole purpose is to clean the raw ASR transcripts outputted by speech-to-text models.<br>Thoroughly tested<br>We evaluated S1-mini on a held-out set of 7,519 cases across 104 transcripts, none of which the model saw during training. Token accuracy reaches 94.8%, with a text-edit error rate of just 11.6%. On email-formatted text, it identifies the greeting line correctly 99.3% of the time and the sign-off 97.9% of the time. It also matches the correct output structure (list versus paragraph) 97.6% of the time, and produces exact email addresses in 92% of cases. Output stability is high as well. Fewer than 1% of generations show any degenerate behavior, such as looping or truncation, and the model correctly withholds output 98.6% of the time when no content should be transcribed.<br>S1-Voice<br>S1-Voice is our cloud-hosted speech-to-text model, trained in-house and hosted by us. It transcribes your words up to 46x faster than it took you to speak them, with most dictations under 30s appearing 0.32 seconds after you stop talking.<br>We evaluated S1-Voice across eight datasets that include meeting audio, earnings calls, and spontaneous speech, not just cleanly read sentences. It...