Àkọlé àwòrán

Àwọn Àkọlé

Wá àwọn àkọlé tí a kọ̀ nínú àwọn àkọlé àwọn àkọlé, àwọn àkọlé tí a kọ̀ nínú àwọn àkọlé àwọn àkọlé, àwọn àkọlé tí a kọ̀ nínú àwọn àkọlé àwọn àkọlé.

Ṣàfihàn

A kò ní àwọn ìrànwọ́ TTS nínú ìtàn rẹ̀. Yọ̀ọ̀kan rán wà láti fàyè gba àwọn rẹ̀! Fi Ojú Rẹ̀ pamọ́

Fi Ojú-ẹ̀yàn Hàn Fún Àwọn Àkọsílẹ̀

200 Àwọn àyọkà

Gbé àwọn fáìlì rẹ̀ lọ́wọ́lọ́wọ́ síbẹ̀, tàbí Wá

MP3, WAV, FLAC, OGG, M4A. Max 500 MB (2 GB on paid plans). Up to 60 seconds is scored.

— tàbí àwòrán láti inú míkrófún rẹ̀ —

00:00

Àwọn ìṣàfihàn — 10 àwọn ìṣàfihàn nínú ọjọ̀ kan, kò ní ìṣàfihàn

Bii o ṣe n ṣiṣẹ

Awọn awari jẹ HuBERT-base classifier (Meta AI, MIT-licensed) fine-tuned lori FoR, ASVspoof, ati WaveFake - mẹta ninu awọn ti o tobi julọ ti awọn eniyan-vs-synthetic ọrọ corpora.

Ò fi àwọn ìṣàfarawé àwọn ìṣàmúlò-ètò tí a tí kọ́ nípa àwọn móòdù TTS, àwọn ìṣàmúlò-ètò ìgbọ́n, tàbí àwọn sítàmú AI mìíràn, pamọ́ sí àwọn ìṣàmúlò-ètò ìròyìn. Ló àwọn ààtò náà bí àkànṣé ìròyìn, kò ní àwọn ìṣàmúlò-ètò ìjọba.

Tí a lò yìí

Awọn iroyin ti n ṣayẹwo awọn fidio ti a fi pamọ tabi ti a ko mọ
Àwọn asesewa ààyè-iṣẹ́ tí n ṣàyẹwo àwọn ààyè-iṣẹ́ ìrànwọ́ nígbà tí a bá pàṣẹ
Àwọn ẹgbẹ́ igbẹ́kẹle-ati-ìdárarẹ́ ti n ṣàfihàn ìdárarẹ́ tí a kọ́lúnà
Àwọn ìṣàfihàn àwọn àwọn ààyè-iṣẹ́ àwọn ìṣàfihàn
Àwọn ìṣàfihàn tí n ṣàfihàn àwòrán wọn nípa àwọn ìṣàfihàn

Àwọn ìrísí-lẹ́tà tí a fọwọ́sì

MP3, WAV, FLAC, OGG, M4A

50MB fáìlì kan. Àwọn ìṣísẹ̀ 60 ìṣàfarawé - àwọn àkóónú ìgbà lórí àwọn àkóónú ìgbà.

Àwọn ààyè-iṣẹ́ ìṣàfarawé

Ṣílẹ̀ ọ̀fẹ́, ṣíṣàfihàn tí o bà fẹ́ diẹ́ sii

Àìfẹ́

10 àwọn ìṣàfihàn nínú ọjọ̀ kan
Kẹ́ẹ̀pútà lọ́wọ́lọ́wọ́ sí 60 àwọn ìṣísẹ̀
Probability + confidence score
Kò ní àwọn àwọn àwọn àwọn àwọn àwọn àwọn

Àwọn Àkọ́gbégbé

Àwọn

15,000 àwọn àwòrán ọ̀fẹ̀ nínú ìṣàfilọ́lẹ̀
Àwọn àmì-àṣírí 200 fún ìṣàfihàn kan
Ìtàn ìṣàfihàn
Àwọn ìròyìn ìjánu-ìṣàfilọ́lẹ̀ gbogbó (àwọn ìṣàmúlò-ètò ìṣàfarawé)

Ṣẹ̀dà

Àwọn

Àwọn ìṣàmúlò-ètò GPU ìṣàfarawé
Àwọn ààyè-iṣẹ́ àwọn àwọn ààyè-iṣẹ́
Àwọn ìṣàfihàn Webhook nípa àwọn ìṣẹ̀dá FAKE
Àwọn Ìṣàmúlò-ètò

_Ṣàfikún

Àwọn Àtòjọ-ẹ̀yàn

It analyses an audio clip and tells you the probability that the speech is AI-generated, cloned, or otherwise synthetic. You get a 0-100% score plus a confidence rating, so you can decide whether the recording is trustworthy enough for publication, payment, or evidence.

A HuBERT-base classifier (Meta AI, MIT licensed) fine-tuned lori mẹta ninu awọn nla olokiki corpora ti eniyan-vs-synthetic ọrọ: FoR (Fake-or-Real), ASVspoof, ati WaveFake. Fine-tune ni a tẹjade bi MelodyMachine/Deepfake-audio-detection-V2 lori Hugging Face.

On the public FoR test set the underlying model reports above 95% accuracy, and on ASVspoof-style attacks it consistently flags TTS and voice-cloning output. Accuracy on novel models the classifier has never seen is lower — treat the score as a strong signal, not legal proof.

O lè fi àwọn àkọlé pamọ́ sí 50MB àti àwọn ìsàlẹ̀-ilà ìtàn 60 sekúndù. Àwọn àkọlé tí o jú lọ nípa ìṣàfarawé àwọn àwọn àmì-ìwé lọ́wọ́lọ́wọ́ lọ́wọ́lọ́wọ́ sí àwọn ààyè-ìwé ìṣàfarawé.

Kò. Kẹ́ìpútà náà tí a fi pamọ́ sínú ìfẹ́fẹ́ lórí àwọn sáà GPU wa, àwọn àwọn ìṣàmúlò-ètò tí a fi pamọ́ sí ọ̀rọ̀ kọ̀ǹpútà àwọn ààtò. A kò fi àwọn àkọsílẹ̀, àwọn ìṣàmúlò-ètò, tàbí àwọn ìṣàmúlò-ètò ìṣàfihàn rẹ̀ lórí ìṣàfihàn.

Detection costs 200 characters per clip — the smallest paid operation on the platform. Free accounts can run a few detections per day; signed-in paid users can run them at full speed with priority queue access.

Ìgbà tí o bá jẹ́ pé òyé jẹ́ ohun tí ò wá jú, kò sí ohun tí ò kù jú, nínú ayè tí ò ní àwọn àwòrán. Òkè nípá tí a fi gbá awọń arákùnrin àti àwọń arákùnrin láti fi ẹrì hàn pé a tí a fi pamọ́ lọ́wọ́lọ́wọ́ lọ́wọ́lọ́wọ́ nípá tí a tí wọn lọ́wọ́lọ́wọ́. Ẹ̀gbẹ̀ ìdárawọ́lú nípá tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí a tí

Kò ní - àwọn àkóónú v1 ní àwọn ààtòjútó àwọn ààtòjútó tí a fi pamọ́. Àwọn ààtòjú àwọn móòdù (iníní TTS wo ní ṣe yìí?) ní pàtó nínú àkànṣe àwọn ààtòjútó àwọn móòdù nígbà tí a bá ní àwọn ààtòjú àwọn móòdù tí a tí àkọ́kọ́.

5.0/5 (1)

Tí o bá fẹ́ bẹrẹ?

Ṣẹ̀dá ọ̀fẹ́ ki o si gba 50 kọ́rẹ́dì. A kò nilò kaadí ẹ̀yàn.

Ṣàfihàn Wó Àwọn Ìtàn

Àwọn Àkọlé

Fi Ojú-ẹ̀yàn Hàn Fún Àwọn Àkọsílẹ̀

Bii o ṣe n ṣiṣẹ

Tí a lò yìí

Àwọn ìrísí-lẹ́tà tí a fọwọ́sì

Àwọn ààyè-iṣẹ́ ìṣàfarawé

Àwọn Àtòjọ-ẹ̀yàn

Kini o n ṣe nígbà tí a bá fi gbọ́ pé o tí ìṣàfilọ́lẹ̀?

Àwọn àwọn ìṣàmúlò-ètò wo nínú àwọn ààyè-iṣẹ́?

How accurate is the result?

Igba wo ni àwòrán mìíràn lè jẹ́?

Àwọn àwọn àwòrán mìíràn tí a fi pamọ́ láti ìṣàfihàn?

Igba wo ni o n lọ?

Kini idi ti eyi fi wa ni ààyè-iṣẹ́ TTS?

O lè sọ nípa ìṣàmúlò-ètò TTS tí o ti ṣẹ̀dà àwọn àyọkà ìsàlẹ̀-ètò?

Tí o bá fẹ́ bẹrẹ?