Model choices
Base and Small prioritize responsiveness. Medium improves accuracy for professional terminology when hardware allows.
Performance
Powerful hardware should feel powerful
Supported GPUs can sharply reduce the wait after speaking. A CPU path remains available. Acceleration changes the compute path, not the intended accuracy target of the selected model.
Performance proof
Less waiting after you speak.
On a reference i9 + RTX 5070 Ti laptop using Whisper Medium and the same test text, EYRO’s GPU-integrated paths averaged 2.27 seconds versus 6.84 seconds without GPU acceleration, about 3.02× faster processing in this internal test.
Non-GPU reference6.84 s
GPU accelerated2.27 s
Results are from an internal founder benchmark. Performance varies by hardware, model, audio length, backend, power mode, and warm/cold state.
Calculated effect
3.02×
faster processing in this internal test
Whisper Medium, same text, i9 + RTX 5070 Ti, two runs per path
Guidance
What changes the wait after you speak
CPU fallback
GPU acceleration is not required. EYRO keeps a graceful CPU path so lower-power machines still complete local dictation.
Warm workers
Keeping the engine warm between recordings reduces repeated cold-start cost on supported setups.
Method
Publishable methodology
Claims stay adjacent to method. The internal reference result used on this site follows these rules.
- Fix audio file, model hash, EYRO version, runtime, language, and decoding settings
- Record OS, CPU, GPU, RAM, power mode, drivers, and plugged-in state
- Separate cold first run from warm steady-state runs
- Prefer median and P95 over fastest-run marketing when sample size allows
- Compare transcript identity / WER, not speed alone, when changing backends
Results are from an internal founder benchmark. Performance varies by hardware, model, audio length, backend, power mode, and warm/cold state.