Your search
Results 8 resources
-
The Fake-or-Real (FoR) dataset is a collection of more than 195,000 utterances from real humans and computer generated speech. The dataset can be used to train classifiers to detect synthetic speech. The dataset aggregates data from the latest TTS solutions (such as Deep Voice 3 and Google Wavenet TTS) as well as a variety of real human speech, including the Arctic Dataset...
-
This repository introduces: π ShiftySpeech: A Large-Scale Synthetic Speech Dataset with Distribution Shifts π₯ Key Features 3000+ hours of synthetic speech Diverse Distribution Shifts: The dataset spans 7 key distribution shifts, including: π Reading Style ποΈ Podcast π₯ YouTube π£οΈ Languages (Three different languages) π Demographics (including variations in age, accent, and gender) Multiple...
-
CodecFake: Enhancing Anti-Spoofing Models Against Deepfake Audios from Codec-Based Speech Synthesis Systems TL;DR: We show that better detection of deepfake speech from codec-based TTS systems can be achieved by training models on speech re-synthesized with neural audio codecs. This dataset is released for this purpose. See our paper and Github for more details on using our...
-
The In-the-Wild dataset contains real and synthetic speech recordings of 58 celebrities and politicians, collected from online videos. It provides a realistic benchmark for testing how well audio deepfake detection models generalize beyond laboratory data such as ASVspoof. Task: Audio Classification (Deepfake / Genuine) Languages: English Modality: Audio Size: 37.9 hours total 17.2 hours fake 20.7 hours real
-
SpeechFake is a large-scale multilingual dataset for speech deepfake detection, featuring over 3 million fake samples across 46 languages. Generated using 30 diverse open-source models* spanning text-to-speech (TTS), voice conversion or clone (VC), and neural vocoder (NV) methods, it offers rich metadata and strong coverage of modern generation techniques, enabling robust and generalizable detection research.
-
PhonemeDF is a large-scale phoneme-level parallel dataset of real and synthetic speech (approximately 730 hours), designed for audio deepfake detection and speech naturalness evaluation. The dataset consists of real speech samples derived from a subset of the LibriSpeech corpus (train-clean-100) and corresponding synthetic speech generated using four Text-to-Speech (TTS) systems (MeloTTS,...
-
All existing databases of spoofed speech contain attack data that is spoofed in its entirety. In practice, it is entirely plausible that successful attacks can be mounted with utterances that are only partially spoofed. By definition, partially-spoofed utterances contain a mix of both spoofed and bona fide segments, which will likely degrade the performance of countermeasures trained with...
-
The automatic speaker verification spoofing and countermeasures (ASVspoof) challenge series is a community-led initiative which aims to promote the consideration of spoofing and deepfakes and the development of countermeasures.
Explore
Audio Data
Tags
- audio data
- synthetic speech (8)
- deepfake (6)
- text-to-speech (TTS) (3)
- voice conversion (VC) (2)
- English (2)
- logical access (1)
- physical access (1)
- spoof (1)
- partial spoof (1)
- phone-level alignment (1)
- multi-language (1)
- Japanese (1)
- Mandarin (1)
Resource type
- Dataset (8)