Pella Research vs Sensity AI
Pella Research
pellaresearch.comSensity AI
sensity.aiSide by side
| Figure | Pella Research | Sensity AI |
|---|---|---|
| Price from | — | — |
| Free plan | — | — |
| Free trial | — | Yes |
| Open source | — | — |
| Web | Yes | Yes |
| Windows | — | Yes |
| API | Yes | Yes |
| Self-hosted | — | Yes |
| Plans published | 0 | 0 |
| Spec score · AI Deepfake Detection Tools | 7.2 (#16) | 7.6 (#12) |
| API | The detection API is a REST endpoint that returns calibrated authenticity scores and per-segment analysis, with access granted on request. | The REST API accepts image, video, or audio submissions by direct upload or public URL and returns a multilayer assessment. |
| Headquarters | Amsterdam, Netherlands | Amsterdam, Netherlands |
| Intended users | Pella describes its work as serving platforms, journalists, courts, and researchers investigating whether audio is synthetic. | Sensity describes its intended users as states, organizations, businesses, and individuals seeking protection from malicious deepfakes. |
In their own words
Pella Research
Pella Research studies synthetic speech and builds models for detecting it in audio. Its research examines vocoder artifacts, spectral statistics, and prosodic cues. A browser demo accepts an uploaded or recorded voice clip and returns a verdict without signup; the site says audio is not stored. The REST detection API returns calibrated authenticity scores and per-segment analysis, with access reviewed after a request, typically within 24 hours. Pella offers the API to journalists, platforms, researchers, and institutions, and lists a free tier for accredited journalists and academics. Published evaluation sets track new voice-cloning systems with documented provenance and no train/test leakage. Pella says models are trained and tested with phone-line compression, background noise, and re-encoding. It reports pellav2 results of 95.82% accuracy, 0.959 F1, and 57 ms latency in the August 2026 Podonos benchmark, plus 93.6% accuracy on m4a files with errors leaning toward false positives. pellav2 is described as open weights under the MIT license.
Sensity AI
Sensity AI analyzes images, video and audio for manipulated faces, AI-generated media, synthetic voices, voice cloning and file tampering. Its multilayer assessment examines visual artifacts, acoustic patterns, metadata, behavioral cues and inconsistencies across media types. Users can submit files or URLs through the web platform, analyze multiple items and track results in a dashboard. The REST API accepts image, video or audio by direct upload or public URL and returns an assessment. Direct uploads are limited to 500 MB, and signed upload URLs expire after one hour. Teams can use member roles; deployment options include cloud and on-premise configurations for servers, workstations or USB dongles. KYC workflows can use an SDK and API, and the enterprise offering describes Microsoft Teams video-call analysis. The Teams app captures 30 seconds of audio and video per analysis, and audio results require speech during that window. Sensity describes its forensic reports as reproducible, explainable and traceable for court support. It offers a trial, but pricing is on request and there is no free plan listed.