CF

Trust & Content Safety

ClearFlow DeepSense Content Trust Infrastructure v2.0

Last updated: March 21, 2026

Our Principles

Provenance first. ClearFlow-originated audio has C2PA credentials, SHA-256 chains, and watermarks. Detection only runs on external audio.

Risk signal, never verdict. We report confidence bands and risk indicators — never "this IS a deepfake."

Honest grading. Per-language accuracy grades are published. We do not claim uniform accuracy across all languages.

Safety is free. Pipeline inline scanning costs zero FlowCoins. Safety should never be paywalled.

Detection Layers

DeepSense uses multiple independent analysis layers with Bayesian fusion. No single layer can determine the result alone.

1
Locale-Aware Heuristics<20ms

Spectral analysis calibrated per language. Detects statistical anomalies in frequency distribution.

2
Adversarial Resilience<200ms

Tests audio stability across 5 codec simulations (MP3, Opus, G.711, noise, pitch shift). Real audio is stable; synthetic is erratic.

3
Glottal Physics<80ms

Analyzes Open Quotient, H1-H2 spectral tilt, jitter, shimmer, and Normalized Noise Energy from voiced segments.

4
Phoneme Boundary Timing<60ms

Measures coefficient of variation in segment durations. Natural speech has irregular timing; TTS is regular.

Layers are weighted via Bayesian fusion. Layer disagreement (> 0.4 delta) triggers "inconclusive" rather than a false verdict.

Per-Language Accuracy Grades

Detection accuracy varies significantly by language. Tonal languages and low-resource languages have lower accuracy. These grades are published transparently via our API.

LanguageGradeNotes
English (US/GB)BBest accuracy. Most training data.
Spanish (ES/MX)CGood but dialectal variation reduces accuracy.
Portuguese (BR)CSimilar to Spanish accuracy.
GermanCCompound word prosody adds challenge.
FrenchCLiaison patterns complicate analysis.
HindiDLimited training data. Retroflex consonants.
KoreanDAgglutinative morphology affects timing.
JapaneseDPitch accent patterns reduce Layer 3 accuracy.
ArabicFTonal. H1-H2 and jitter layers disabled.
Mandarin ChineseFTonal. Fundamental frequency carries meaning.
Filipino/TagalogFAustronesian prosody poorly calibrated.
Yoruba / Hausa / Swahili?Ungraded. Insufficient calibration data.

Grades reflect honest self-assessment, not marketing claims. Live data available at GET /api/trust/languages

EU AI Act Article 50 Compliance

[LIVE]

Technical marking

LSB steganographic watermark in every synthesized audio sample

[LIVE]

Machine-readable flag

synthetic: true in every API response + C2PA manifest

[LIVE]

Visible labeling

'AI-generated voice' badge on all playback surfaces

[LIVE]

Provenance persistence

SHA-256 fingerprint + watermark ID stored for 2-year retention

[LIVE]

Public verification

POST /api/trust/provenance-lookup — check any audio hash

[LIVE]

Multi-layer detection

4-layer DeepSense stack with per-locale calibration

Victim Protection Tools

If you believe your voice has been cloned or used without consent, ClearFlow provides forensic tools for evidence generation and takedown support.

Evidence Package

Upload audio to generate a forensic analysis report suitable for legal proceedings and platform takedown requests.

POST /api/trust/victim-verifyAuth + consent required. 5 per day.

Protected Voice Registry

Register your voiceprint hash in our public registry. Other platforms can check against it before allowing voice cloning.

POST /api/trust/protected-voiceAuth + consent required.

Provenance Lookup

Check if an audio hash exists in ClearFlow's provenance chain. Determine if we generated specific content.

POST /api/trust/provenance-lookupPublic. Rate limited.

Trust API

POST
/api/trust/verify

Primary scan endpoint. Returns risk bands (not point estimates). (Auth required)

GET
/api/trust/languages

Published per-language accuracy grades. (Public)

POST
/api/trust/provenance-lookup

Audio hash origin check. (Public (100/hr))

GET/POST/DELETE
/api/trust/consent

Consent management (GDPR Art.7). (Auth required)

POST
/api/trust/victim-verify

Evidence package generation. (Auth + consent)

POST/GET/DELETE
/api/trust/protected-voice

Voiceprint hash registry. (Auth for write)

GET
/api/trust/red-team

Published red team test results. (Public)

Honest Limitations

1. No ML model. All detection uses DSP heuristics. Sophisticated TTS can likely defeat current layers. Architecture is ready for ONNX models in future phases.

2. Accuracy is limited. Grade B for English (best). Grade F for Arabic/Mandarin. Published honestly above.

3. Detection is risk signal, not proof. Legal proceedings need human expert analysis. Our output supports, not replaces, expert judgment.

4. Provenance is the real defense. Detection is an arms race. C2PA + watermarks are the sustainable long-term mechanism.

ClearFlow Inc. Provenance-first AI audio. Every sample watermarked, signed, and auditable.