Deepfake Detector vs Pella Research
Deepfake Detector
deepfakedetector.aiPella Research
pellaresearch.comSide by side
| Figure | Deepfake Detector | Pella Research |
|---|---|---|
| Price from | $4,939/mo | — |
| Free plan | Yes | — |
| Free trial | — | — |
| Open source | — | — |
| Web | Yes | Yes |
| API | Yes | Yes |
| Plans published | 4 | 0 |
| Spec score · AI Deepfake Detection Tools | 8.5 (#4) | 7.2 (#16) |
| API | A public REST API accepts a media URL or file and returns JSON detection results. | The detection API is a REST endpoint that returns calibrated authenticity scores and per-segment analysis, with access granted on request. |
| Headquarters | Montreal, Quebec, Canada | Amsterdam, Netherlands |
Plans
Deepfake Detector
- FreeFree
- Starter$4,939 a month
- BusinessPrice not published
- EnterprisePrice not published
In their own words
Deepfake Detector
Deepfake Detector analyzes audio, video and images, returning an Authentic, Likely Synthetic or Inconclusive verdict with a confidence score. Video analysis works at frame level, and applicable plans support clips up to 10 minutes. Coverage includes video generators such as Sora, Runway, Pika and Kling; image generators such as MidJourney, DALL·E, Stable Diffusion and Flux; and voice platforms including ElevenLabs, PlayHT, Resemble and OpenAI. The service is available on the web, through a Chrome extension and via a public REST API that accepts a media URL or file and returns JSON results. The extension scans webpage media from a right-click menu and works in Chromium browsers including Chrome, Edge, Brave and Arc. For files over 60 seconds, webhooks can return the verdict when analysis completes; type-safe SDKs are available for Python, Node, Go and Ruby. Uploaded files are purged within 60 seconds after a verdict, while hashed identifiers and verdict metadata are retained for billing and audit. A free plan allows 50 detections per month; paid plans start at $49/mo.
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.