A Comprehensive Survey with Critical Analysis for Deepfake Speech Detection

Table II: THE CHALLENGE COMPETITIONS PROPOSED FOR DEEPFAKE SPEECH DETECTION

Challenge Competitions	Years	Data type	Languages	Public Label (train&dev/test)	Audio	Visual	Team No.	Top-1 System
ASVspoof 2015 (audio) [15]	2015	Speech	English	Yes/Yes	Yes	No	16	Ensemble
ASVspoof 2019 (LA Task) [16]	2019	Speech	English	Yes/Yes	Yes	No	48	Ensemble
DFDC [17], [18]	2020	Speech	English	Yes/Yes	Yes	Yes	2114	Ensemble
FTC [19]	2020	Speech	English	No/No	Yes	No	n/a	n/a
ASVspoof 2021 (LA Task) [20]	2021	Speech	English	Yes/Yes	Yes	No	41	Ensemble
ASVspoof 2021 (DF Task) [20]	2021	Speech	English	Yes/Yes	Yes	No	33	Ensemble
ADD 2022 Track 1 [21]	2022	Speech	Chinese	Yes/Yes	Yes	No	48	Single model
ADD 2022 Track 2 [21]	2022	Speech	Chinese	Yes/Yes	Yes	No	27	Single model
ADD 2022 Track 3.2 [21]	2022	Speech	Chinese	Yes/Yes	Yes	No	33	Single model
ADD 2023 Track 1.2 [22]	2023	Speech	Chinese	No/No	Yes	No	49	Ensemble model
ADD 2023 Track 2 [22]	2023	Speech	Chinese	No/No	Yes	No	16	Single model
AV-Deepfake1M [23], [24]	2024	Speech	English	Yes/No	Yes	Yes	n/a	n/a
ASVspoof 2024 [25]	2024	Speech	English	Yes/No	Yes	No	53	Ensemble model
SVDD 2024 [26], [27]	2024	Singing	Multi-language (6)	Yes/No	Yes	No	47	Ensemble model

Table III: PUBLIC AND BENCHMARK DATASETS PROPOSED FOR DEEPFAKE SPEECH DETECTION

Dataset	Year	Language	Speakers (Male/Female)	Utt. No. (Real/Fake)	AI-Synthesized Speech Systems	Speech Condition	Real Speech Resources	Utt. length	Evaluation Metrics
ASVspoof 2015 [15](audio)	2015	English	45/61	16,651/246,500	10	Clean	Speaker Volunteers	1 to 2	EER
FoR [30](audio)	2019	English	33	198,000+	7	Clean	Kaggle [31]	2.35	Acc
ASVspoof 2019 (LA task) [16](audio)	2019	English	46/61	121,461	19	Clean & Noisy	Speaker Volunteers	n/a	EER
DFDC [32](video)	2020	English	3426	12,8154/104,500	1	Clean & Noisy	Speaker Volunteers	68.8	Precision/ Recall
ASVspoof 2021 (LA task) [20](audio)	2021	English	21/27	18,452/163,114	13	Clean & Noisy	Speaker Volunteers	n/a	EER
ASVspoof 2021 (DF task) [20](audio)	2021	English	21/27	22,617/589,212	100+	Clean & Noisy	Speaker Volunteers	n/a	EER
WaveFake [33](audio)	2021	English, Japanese	0/2	117,985	6	Clean	LJSPEECH [29] & JSUT [30]	6s/4.8s	EER
KoDF [36](video)	2021	Korean	198/205	62,116/175,776	2	Clean	Speaker Volunteers	90/15 (real/fake)	Acc & AuC
ADD 2022 [21]	2022	Chinese	40/40	3012/24072	2	Clean	AISHELL-3 [37]	1s to 10s	EER
FakeAVCeleb [38](video)	2022	English	250/250	570/25000	2	Clean & Noisy	Vox-Celeb2 [39]	7s	AuC
In-the-Wild [40](audio)	2022	English	58	31779	0	Clean & Noisy	Self-collected	4.3s	EER
LAV-DF [41](video)	2022	English	153	36,431/99,873	1	Clean & Noise	Vox-Celeb2 [39]	3s to 20s	AP
Voc.v [42](audio)	2023	English	46/61	82,048	5	Clean & Noisy	ASVspoofing 2019 LA	n/a	EER
CFAD [43](audio)	2023	Chinese	1023	374,000	12	Clean & Noisy & Codecs	AISHELL1-3 [44], [45], MAGICDATA [46]	n/a	EER
PartialSpoof [47](audio)	2023	English	46/61	12,483/108,87	19	Clean & Noisy	ASVspoofing 2019	0.2s-6.4s	EER
LibriSeVoc [48](audio)	2023	English	n/a	13,201/79,06	6	Clean & Noisy	Librispeech	5s-34s	EER
AV-Deepfake1M [23], [24](video)	2023	English	2,068	286,721/860,039	2	Clean & Noisy	Vox-Celeb2 [33]	5s-35s	Acc & AuC
MLAAD [49](audio)	2024	Multi-Language(23)	n/a	76,000	54	Clean & Noisy	M-AILABS [50]	n/a	Acc.
ASVspoof 2024 [25](audio)	2024	English	964/958	188,819/815,262	28	Clean & Noisy	MLS [51]	n/a	EER
SVDD2024 [26](audio)	2024	Multi-Language (6)	59	12,169/72,235	48	Clean	Mandarin & Japanese [27]	n/a	EER

Table IV: DEEPFAKE SPEECH GENERATION SYSTEMS USED IN PUBLIC DSD DATASETS (TTS: TEXT TO SPEECH, VC: VOICE CONVERSION, AT: ADVERSARIAL ATTACH USING MALAFIDE OR MALOCOPULA)

Datasets	Year	No. of TTS/VC/AT	Deepfake Speech Generation Systems
ASVspoof 2015 [15]	2015	7 VC, 3 TTS	VC-01 [52], [53], VC-02 [54], TTS-01 [55], TTS-02 [55], VC-03 [56], VC-04 [57], VC-05 [57], VC-06 [58], VC-07 [59], TTS-03 [60]
FoR [30]	2019	7 TTS	Deep Voice 3, Amazon AWS Polly, Baidu TTS, Google Traditional TTS, Google Cloud TTS, Google Wavenet TTS, Microsoft Azure TTS
ASVspoof 2019 (LA task) [16]	2019	8 VC, 11 TTS	TTS-01 [61], TTS-02 [61], [62], TTS-03 [63], TTS-04 [64], VC-01 [65], VC-02 [66], TTS-05 [63], [67], TTS-06 [61], [68], TTS-07 [69], [70], TTS-08 [71], [72], TTS-09 [71], [72], [73], TTS-10 [74], VC-03+TTS [75], VC-04+TTS [76], [77], VC-05+TTS [76], [77], TTS-11 [64], VC-06 [78], [79], VC-07 [80], [81], [82], VC-08 [66]
DFDC [32]	2020	1 TTS	TTS Skins voice conversion [83]
KoDF [36]	2021	2 TTS	ATFHP [84] and Wav2Lip [85]
ASVspoof 2021 (LA task) [20]	2021	13 TTS/VC	Reuse ASVspoof 2019
ASVspoof 2021 (DF task) 20]	2021	100 TTS/VC	Vocoders [86]
WaveFake [33]	2021	6 TTS	MelGAN [87], FB-MelGAN [87], HiFi-GAN [88], WaveGlow [89], PWG [90], MB-MelGAN [87]
FakeAVCeleb [38]	2022	2 TTS	SV2TTS [91], [92]
In-the-Wild [40]	2022	n/a	n/a
LAV-DF [41]	2022	1 TTS	SV2TTS [93]
Voc.v [42]	2023	5 TTS	HiFi-GAN [88], MB-MelGAN [87], WaveGlow [89], PWG [90], Hn-NSF [94]
CFAD [43]	2023	11 TTS	STRAIGHT [95], Griffin-Lim [96], LPCNet [97], WaveNet [98], PWG [90], HiFi-GAN [99], MB-MelGAN [87], MelGAN [87], WORLD [100], FastSpeech [101], Tacotron-HifiGAN [102]
PartialSpoof [47]	2023	21 TTS/VC	Reuse ASVspoof 2019
LibriSeVoc [48]	2023	6 TTS/VC	WaveNet [98], WaveRNN [103], MelGAN [87], Parallel WaveGAn [104], WaveGrad [105], DiffWave [106]
AV-Deepfake1M [23], [24]	2023	2 TTS	VITS [107], YoursTTS [108]
MLAAD [49]	2024	54 TTS	Bark, Capacitron, FastPitch, GlowTTS, Griffin Lim, Jenny, NeuralHMM, Overflow, Parler TTS, Speech5, Tacotron DDC, Tacotron2, Tacotron2 DCA, Tacotron2 DH, Tcotron2-DDC, Tortoise, VITS, VITS Neon, VITS-MMS, XTTS v1.1, XTTS v2
ASVspoof 2024 [25]	2024	15 TTS, 6 VC, 7 AT	TTS-01 [109], TTS-02 [110], TTS-03 [111], TTS-04 [112], TTS-05 [113], TTS-06 [114], TTS-07 [115],TTS-08(self-develop), VC-01 [116], TTS-09 [117], VC-02 [118], VC-03(self-develop), TTS-10 [119], AT-01 (Malafide+TTS-10 [119]), TTS-11 [120], AT-02(self-Develop), TTS-12 [121], TTS-13 [122], AT-03(Malafide+TTS [123]), VC-04(self-develop), VC-05 [124], VC-06(add noise), AT-04(Malacopula+VC-06), TTS-14 [125], TTS-15 [126], AT-05(Malacopula+AT-01), AT-06(Malacopula+TTS-13 [122]), AT-07(Malacopula+VC-05 [124])