Academic / research access only

Requests are reviewed manually. Incomplete requests, or personal-email requests without a detailed justification, will be denied.

This model is released under CC-BY-NC-4.0 for non-commercial academic and research use only. Institutional/university email addresses are strongly preferred — requests from personal email domains (e.g. gmail, outlook) need a more detailed explanation of intended use and are approved only in rare cases. By requesting access you agree to acknowledge this work in any resulting publication, model, or dataset that uses it for evaluation or training (see the Citation section below).

forensics_0.3B_v2_deepfake_age_gender_classifier

Deepfake detection + speaker age & gender - hardened against the newest open-source TTS systems

🔒 Gated — access requires approval. Latest release — Q3 2026.

The newest, most advanced model in the Forensics family. Purpose-built and tested against the latest open-source TTS/voice-synthesis engines — qwen3, voxcpm, and omnivoice — plus AI-generated music as an additional adversarial signal, this model achieves state-of-the-art results against the latest open-source TTS models: 99%+ accuracy on the held-out test set of each. This is the sharpest edge of the Forensics family against next-generation synthetic voice.

On top of real/fake detection, it estimates the speaker's age and gender in the same forward pass — one model, three outputs.

Access to this repo requires manual approval. Submit a request from the repo page and it'll be reviewed directly.

Backbone: microsoft/wavlm-large (~300M params, fully unfrozen)
Pooling: AASIST graph-attention
Extra heads: age (regression) + gender (female / male / child)

Part of the Forensics family

Model	Use it for
`forensics_0.3B_base_deepfake_classifier`	general-purpose default
`forensics_0.3B_xlsr_wild_deepfake_classifier`	uncontrolled / real-world audio
`forensics_0.3B_v2_deepfake_age_gender_classifier` 🆕 (this model, gated)	speaker age/gender, hardened against the newest TTS threats — our latest release
`forensics_0.3B_wavlm_oc_softmax_deepfake_classifier`	tighter bonafide boundary, ensembling

Full family: huggingface.co/collections/eliya/forensics-speech-deepfake-detection-family

Training

Trained using an agentic training loop — see eliyasegev/autotrain.

Multi-task objective: cross-entropy + OC-Softmax + supervised contrastive for deepfake detection, plus a regression loss for age and a classification loss for gender — all trained jointly.

Trained on 5M+ speech samples spanning a mix of proprietary and publicly available data — some of our own self-created data has been open-sourced as well. Built for robustness against a wide range of TTS generation methods — including some of the most advanced synthesis techniques available today, plus the newest TTS engines (qwen3, voxcpm, omnivoice). Also trained on environmental sound and music, extending fake-detection beyond speech to those domains as well. Every row additionally carries age/gender labels. AdamW, cosine LR schedule, and the full Forensics augmentation stack (codec transcoding, noise augmentation, RIR, RawBoost, SpecAugment, FreqMask, splice/mix, cross-class splice).

Results

Deepfake-detection EER across the standard external benchmark suite:

Eval set	EER %
Val (held-out)	1.11
MLAAD (v7)	0.71
CodecFake	0.81
DFADD	0.00
MD-CommonVoice	0.35
In-the-Wild	2.20
ASVspoof2019-LA	0.85
ASVspoof2021-LA	3.49
ASVspoof2024	14.09
ADD2022-Track1	16.86
ADD2022-Track3	2.88
ADD2023-Round1	5.53
ADD2023-Round2	11.60
LibriSeVoc	0.08
SONAR	1.14
Avg (all sets)	4.11
Avg (external only)	4.33

On the three newest-TTS test sets (qwen3, voxcpm, omnivoice) — the threats this model was specifically hardened against — accuracy exceeds 99%. Age/gender evaluation plots are included alongside the model weights.

Files in this repo

file	purpose
`checkpoint_epoch_9.safetensors`	model weights, safe format
`checkpoint_epoch_9.pt`	model weights, legacy pickle
`config.json`	minimal architecture metadata (also used by the Hub to track downloads)
`inference.py`	run script — prefers the `.safetensors` file automatically
`model.py`	architecture
`requirements.txt`	deps

Setup

pip install -r requirements.txt   # torch, torchaudio, transformers, safetensors
hf download eliya/forensics_0.3B_v2_deepfake_age_gender_classifier --local-dir .

(This is a gated repo — you'll need an approved access request before the download works.)

Run

python inference.py <audio.wav>

(Optionally override the checkpoint: python inference.py <audio.wav> <checkpoint.pt>.)

Audio is auto-converted to mono / 16 kHz and trimmed/padded to 5 s.

Output

fake_probability: <0..1>      # threshold is domain-dependent — adjust to your use case; ~0.1-0.2 is usually the best range
bonafide_score:   <0..1>      # raw P(real)
verdict: REAL | FAKE
age_years: <n>
gender: female | male | child  (conf ..)

Example

$ python inference.py real_human.wav
fake_probability: 0.1022
bonafide_score:   0.8978
verdict: REAL

$ python inference.py designed_tts_voice.wav
fake_probability: 0.8646
bonafide_score:   0.1354
verdict: FAKE
age_years: 34
gender: male  (conf 0.98)

Higher fake_probability = more likely a deepfake. Score is 1 − sigmoid(logit), since the classifier is trained with label 1 = real, 0 = fake.

References

WavLM: Chen et al., 2022, "WavLM: Large-Scale Self-Supervised Pre-Training for Full Stack Speech Processing", arXiv:2110.13900
AASIST: Jung et al., 2021 (ICASSP 2022), "AASIST: Audio Anti-Spoofing Using Integrated Spectro-Temporal Graph Attention Networks", arXiv:2110.01200
OC-Softmax: Zhang et al., 2020 (IEEE SPL 2021), "One-Class Learning Towards Synthetic Voice Spoofing Detection", arXiv:2010.13995
Supervised Contrastive Learning: Khosla et al., NeurIPS 2020, arXiv:2004.11362
Age/gender head design (age regression scaled to 0-100 years, 3-class female/male/child gender): Burkhardt et al., 2023, "Speech-based Age and Gender Prediction with Transformers", arXiv:2306.16962
CCC loss (age regression): Lin, 1989, "A Concordance Correlation Coefficient to Evaluate Reproducibility", Biometrics
RawBoost augmentation: Tak et al., 2021 (ICASSP 2022), arXiv:2111.04433
FreqMask augmentation: Xie et al., 2024, arXiv:2408.06922
SpeechFake dataset: ACL 2025, "SpeechFake: A Large-Scale Multilingual Speech Deepfake Dataset Incorporating Cutting-Edge Generation Methods", arXiv:2507.21463

License

CC-BY-NC-4.0 — free for personal and research use.

Citation

If you use this in research, evaluation, or training, please acknowledge it:

@misc{segev2026forensicsv2,
  author       = {Segev, Eliya},
  title        = {Forensics 0.3B v2: Speech Deepfake Detection + Age/Gender Classifier},
  year         = {2026}
}

Downloads last month: -

Model tree for eliya/forensics_0.3B_v2_deepfake_age_gender_classifier

Base model

microsoft/wavlm-large

Finetuned

(39)

this model

Datasets used to train eliya/forensics_0.3B_v2_deepfake_age_gender_classifier

Collection including eliya/forensics_0.3B_v2_deepfake_age_gender_classifier

Forensics - Speech Deepfake Detection Family

Collection

WavLM/XLS-R + AASIST speech deepfake detectors. Free for personal and research use (CC-BY-NC-4.0) - contact eliya@vocos.io for commercial licensing. • 4 items • Updated Sep 1 • 1

Papers for eliya/forensics_0.3B_v2_deepfake_age_gender_classifier

SpeechFake: A Large-Scale Multilingual Speech Deepfake Dataset Incorporating Cutting-Edge Generation Methods

Paper • 2507.21463 • Published Jul 29, 2025

Temporal Variability and Multi-Viewed Self-Supervised Representations to Tackle the ASVspoof5 Deepfake Challenge

Paper • 2408.06922 • Published Aug 13, 2024

Speech-based Age and Gender Prediction with Transformers

Paper • 2306.16962 • Published Jun 29, 2023

RawBoost: A Raw Data Boosting and Augmentation Method applied to Automatic Speaker Verification Anti-Spoofing

Paper • 2111.04433 • Published Nov 8, 2021

WavLM: Large-Scale Self-Supervised Pre-Training for Full Stack Speech Processing

Paper • 2110.13900 • Published Oct 26, 2021 • 3