---
canonical: https://biosonic.io/benchmarks
---

# 3rd party validations of BioSonic Accuracy.

Last updated: 2026-07-20.

Third-party tests · independent validation.

Three independent tests by three separate teams. Same conclusion every time.

[Book a demo](https://cal.com/biosonic/30min) | [Try for free](https://bat.biosonic.se/signup)

Independent benchmarks prove BioSonic's bat sound analysis leads the field. 98.7% F1 score vs 73% for Kaleidoscope Pro. 13.4× fewer false positives.

The benchmarks on this page are labelled:

- 01 · Wilder Sensing · third-party · F1 accuracy
- 02 · Jan Drachmann · field · 5 detectors · 11 species

## Benchmark 01 · Third-party testing — Independent validation: 98.7% F1 accuracy.

Wilder Sensing · Geoff Carss + Annabel Jeffries

**98.7% F1 ACCURACY**

> "We tested a range of existing solutions using real acoustic data, and BioSonic clearly stood out, achieving 98.7% accuracy (F1 score) and giving us full confidence in the results we're bringing to the platform."
> — Geoff Carss & Annabel Jeffries, Wilder Sensing

### BioSonic's AI model uses Image recognition on Spectrogram. This leads to a step change in accuracy.

Grassland + Paludiculture dataset · 5,325 predictions.

#### Metric comparison — Precision, Recall, F1

Higher is better · bars start at 50% baseline.

| Metric | BioSonic | Kaleidoscope | BioSonic advantage |
|---|---|---|---|
| Precision | 97.8% | 61.4% | +36.4pt |
| Recall | 99.6% | 91.7% | +7.9pt |
| F1 score | 98.7% | 73.0% | +25.7pt |

#### BioSonic · Confusion Matrix · Accuracy 99.7%

| | Predicted bat | Predicted no bat |
|---|---|---|
| **Actual bat** | 730 (true positive) | 3 (false negative) |
| **Actual no bat** | 13 (false positive) | 4,579 (true negative) |

#### Kaleidoscope Pro · Confusion Matrix · Accuracy 95.8%

| | Predicted bat | Predicted no bat |
|---|---|---|
| **Actual bat** | 543 (true positive) | 49 (false negative) |
| **Actual no bat** | 174 (false positive) | 4,503 (true negative) |

#### False positives (wrong "bat" calls)

Cost: investigator time wasted reviewing non-bat audio, inflated species counts.

| Classifier | False positives |
|---|---|
| BioSonic | 13 |
| Kaleidoscope | 174 |

Kaleidoscope: 13.4× more.

#### False negatives (missed bats)

Cost: protected species overlooked, incomplete EIA, compliance risk.

| Classifier | False negatives |
|---|---|
| BioSonic | 3 |
| Kaleidoscope | 49 |

Kaleidoscope: 16.3× more.

Same data, very different outcomes. BioSonic delivers 98.7% F1 with only 16 errors across 5,325 predictions. Kaleidoscope Pro misses nearly 9% of true bats and produces 13× more false calls.

So BioSonic's F1 accuracy is roughly 25 percentage points higher. The most striking difference is precision: BioSonic was correct 97.8% of the time it flagged a species, vs 61.4% for Kaleidoscope. That precision gap is what drives the validation-time savings, since Kaleidoscope produced far more false positives to weed out (277 vs 16).

## Benchmark 02 · Southern Denmark — Manual vs BioSonic.

By Jan Drachmann · independent biologist · 5 detectors, 11 species · manual analysis vs BioSonic, same dataset.

### Detector ranking — Identical

D4 > D5 > D2 > D1 > D3 — same order whether counted by hand or by AI. Same ecological conclusion about which sites had the most activity.

### Detections per detector

Same order, slightly higher counts. "JD" is Jan Drachmann's manual count; "BS" is BioSonic.

| Detector | Manual (JD) | BioSonic (BS) |
|---|---|---|
| D4 | 23,527 | 25,514 |
| D5 | 2,370 | 3,505 |
| D2 | 1,156 | 1,688 |
| D1 | 391 | 461 |
| D3 | 94 | 118 |

### Natter's bat calls: 11 → 100 — Jan found 11. BioSonic found 100.

Going through tens of thousands of files by hand, it's easy to miss a few. On review, every one of BioSonic's 100 Natter's bat calls was validated as correct. That's the payoff of AI on large datasets — rare species don't slip through.

### Species breakdown — Common & Rare species comparison

#### Common Species

BioSonic detects more Pipistrel (+203%) and Trold (+16%).

| Species (as labelled on the page) | Manual | BioSonic |
|---|---|---|
| Dværg | 13,556 | 10,669 |
| Trold | 11,433 | 13,248 |
| Pipistrel | 2,257 | 6,834 |

#### Rare & Protected Species

Separate scale — these are the ones that matter for Annex II compliance.

| Species (as labelled on the page) | Manual | BioSonic |
|---|---|---|
| Vand | 112 | 221 |
| Frynse | 11 | 100 |
| Dam | 0 | 0 |
| Brun | 40 | 41 |
| Syd | 55 | 67 |

### Weather integration

Jan also found interesting bat behaviours compared to wind speed and rain with BioSonic's automatic weather integration. The page shows two charts, switchable between **Wind** (bat activity vs wind speed) and **Rain** (bat activity vs rain).

## And the human cost — 2TB bat data, comparing BioSonic vs Kaleidoscope.

According to a consultancy in Southern England.

| Workflow | Analysis time | Time to client |
|---|---|---|
| Kaleidoscope workflow | 490 hrs | 12 weeks |
| BioSonic workflow | 163 hrs | 4 weeks |

✓ 327 hours back · This means £21,255 saved at standard UK consultant rate.

## Watch — BioSonic on BatAbility Club with Neil Middleton.

The page embeds the BatAbility Club session with Neil Middleton discussing BioSonic.

## Run it on your own data — A 30-minute demo. Bring a WAV, leave with the report.

[Book a demo](https://cal.com/biosonic/30min) | [Try for free](https://bat.biosonic.se/signup)

josef.carlson@biosonic.se · +46 72 744 65 85

## Sources

- **Benchmark 01** — Wilder Sensing, Geoff Carss and Annabel Jeffries. Third-party test on the Grassland + Paludiculture dataset, 5,325 predictions. BioSonic 98.7% F1 (precision 97.8%, recall 99.6%, overall accuracy 99.7%) vs Kaleidoscope Pro 73.0% F1 (precision 61.4%, recall 91.7%, overall accuracy 95.8%).
- **Benchmark 02** — Jan Drachmann, independent biologist, southern Denmark. Manual analysis vs BioSonic on the same dataset: 5 detectors, 11 species.
- **Time-and-cost figures** — a consultancy in southern England, 2TB of bat data: 490 hrs / 12 weeks with Kaleidoscope vs 163 hrs / 4 weeks with BioSonic; 327 hours saved, £21,255 at standard UK consultant rate.
- **Video** — BioSonic on BatAbility Club with Neil Middleton.
- All numbers on this page are reproduced exactly as published on https://biosonic.io/benchmarks.

Last verified: 2026-07-20.

## Related Pages

- [Home](https://biosonic.io/)
- [BioSonic vs Kaleidoscope Pro](https://biosonic.io/compare/kaleidoscope) — the full head-to-head
- [Compare bat analysis software](https://biosonic.io/compare)
- [Product](https://biosonic.io/product)
- [Customers](https://biosonic.io/customers) · [Wharton](https://biosonic.io/wharton) · [DMR](https://biosonic.io/dmr) · [COWI](https://biosonic.io/cowi)
- [FAQ](https://biosonic.io/faq)
- [Automatic bat species identification](https://biosonic.io/automatic-bat-species-identification)
- [Wind farm bat monitoring](https://biosonic.io/wind-farm-bat-monitoring)
- [Free tools](https://biosonic.io/tools) · [Time calculator](https://biosonic.io/tools/time-calculator)