Pella Research vs Tayanch
Pella Research
pellaresearch.comTayanch
tayanch.comSide by side
| Figure | Pella Research | Tayanch |
|---|---|---|
| Price from | — | $249/mo |
| Free plan | — | Yes |
| Free trial | — | — |
| Open source | — | — |
| Web | Yes | Yes |
| API | Yes | Yes |
| Plans published | 0 | 6 |
| Spec score · AI Deepfake Detection Tools | 7.2 (#16) | 7.8 (#10) |
| API | The detection API is a REST endpoint that returns calibrated authenticity scores and per-segment analysis, with access granted on request. | The API uses REST over HTTPS, JSON responses, and API-key authentication through the X-API-Key header; it accepts files and media links. |
| Headquarters | Amsterdam, Netherlands | United Arab Emirates |
Plans
Tayanch
- StarterFree
- Pay as you go — image check$0.01, paid once
- Pay as you go — audio check$0.03, paid once
- Pay as you go — video check$0.08, paid once
- Team$249 a month
- EnterprisePriced on request
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.
Tayanch
Tayanch analyzes images and video to assess whether media is real, AI-generated, or uncertain, and returns probability and confidence information. Image checks can identify fully generated pictures and use a heat map to locate AI edits in a real photo. Video analysis covers face manipulation and provides a frame-level advisory for generated footage. Audio analysis for voice clones and synthetic speech is under maintenance and coming soon. Ambiguous or heavily processed media can receive an uncertain verdict rather than a forced classification. Detection combines vision-transformer embeddings with camera-physics and compression checks, while Content Credentials and generator markers are checked first. Its REST API uses HTTPS, JSON responses, and API-key authentication through the X-API-Key header; it accepts files and media links. A free API key includes 100 checks and allows 30 requests per minute without a card. Image checks cost $0.01 as a one-time purchase, and video checks cost $0.08 as a one-time purchase. Tayanch says submitted media is analyzed in memory, discarded after the verdict, and not used for training. It cautions that results are probabilistic evidence, not a sole basis for accusations or consequential action. The self-serve API has no SLA; an SLA is available only through a signed enterprise agreement.