Choosing the Right AI Model

Available Models

Weesper Neon Flow includes seven local recognition profiles. The download size is a useful planning measure; actual memory use also depends on your operating system and workload.

ProfileDetail LevelSpeedDownload SizeLanguage CoverageBest For
MinimalEverydayVery fast~75 MBMultilingualOlder hardware and quick notes
CompactEverydayVery fast~148 MBMultilingualLightweight daily dictation
Compact+ (EN)English-optimisedVery fast~145 MBEnglish onlySharper everyday English transcription
EverydayDetailedFast~190 MBMultilingualFast everyday writing
StandardDetailedFast~512 MBMultilingualMost users (recommended default)
PrecisionHigh detailBalanced~547 MBMultilingual transcriptionPolished notes, accents, and technical terms in the spoken language
ProfessionalHigh detailSteady~3.0 GBMultilingualAccuracy-first professional work

Minimal (EN), Extended, and Expert are legacy profiles retained for existing installations. They are not offered for new downloads. Compact+ (EN) is the current English-only profile; it is not suitable when you dictate in another language. Precision transcribes the spoken language and does not translate to English.

How to Change Models

  1. Open SettingsRecognition Quality
  2. Select your preferred profile
  3. The model loads automatically for your next dictation

Recommendations

Understanding Transcription Speed

When you release the hotkey, there is a short delay before your text appears. This is completely normal — Weesper Neon Flow is transcribing your audio locally on your device, and the time it takes depends on three factors:

FactorImpact on SpeedImpact on Accuracy
Recognition profile (Minimal → Professional)Larger profiles are slowerLarger profiles are more accurate
Accuracy levelHigher accuracy takes longerBetter transcription quality
Your hardware (CPU, RAM)Faster hardware = faster resultsNo impact

The Speed–Quality Tradeoff

Think of it as a triangle: model power, accuracy, and hardware. You control the first two — adjust them to match your hardware:

Finding Your Sweet Spot

If transcription feels too slow after you release the hotkey:

  1. Go to SettingsRecognition Quality
  2. Try a lighter profile first — switch from Professional or Precision to Standard, Everyday, Compact, or Minimal
  3. Then adjust accuracy — lower it one step and test again
  4. Repeat until you find the right balance between speed and quality

The best configuration is the one where you barely notice the delay while still getting accurate transcriptions. Test with a few short sentences after each change.

Shorten Long Pauses

Enable Shorten long pauses in Settings → Recognition Quality to make long dictations faster by trimming silence before transcription.

FAQ

Which model should I choose?

Standard is the recommended multilingual default for most users. Use Minimal or Compact for older hardware, Everyday for fast daily writing, Precision for polished multilingual transcription in the language you speak, and Professional when accuracy matters most. Minimal (EN), Extended, and Expert are legacy profiles retained for existing installations and are not offered for new downloads.

Can I switch models?

Yes, go to Settings → Recognition Quality to switch profiles at any time.

Why is there a delay before the text appears?

After you release the hotkey, Weesper transcribes your audio locally on your device. The time this takes depends on three factors: the AI model (larger = slower but more accurate), the accuracy level, and your hardware power (CPU, RAM). This is normal — try a lighter model or lower accuracy level if the delay is too long.

What is the best combination of model and accuracy for my computer?

Start with Standard and the Everyday quality setting. If it feels slow, try Compact or Minimal. If accuracy matters more than speed, try Precision or Professional.

Was this article helpful?

Still need help? Contact support

support@weesperneonflow.ai
Ready to supercharge your workflow? Download Free Trial