Tevere: AI Agents Mimhanzi Kutaura-Ku-Tekisi Voice Cloning Kutaura Marketplace More
Kuzivisa Bug / Feature Request

AI Lip Sync Video Generator

Upload a face photo and an audio clip — get a talking-head video with realistic lip sync, head pose, and blinks. Powered by SadTalker (MIT). Commercial use OK.

Sign Up Free
Hatina TTS mazwi muchirungu chako parizvino. Tibatsire kuti tigadzirise ako! Kutengesa mashoko ako

Kuisa Face + Audio

1,000 characters per second

Drag & drop your file here, or browse

JPG, PNG, or short MP4/WebM. Max 10MB. One clear, well-lit face works best.

file.mp3

0 MB

Drag & drop your file here, or browse

MP3, WAV, M4A, or FLAC. Max 10MB. Free: up to 30 sec. Pro: up to 5 min.

file.mp3

0 MB

Rendering yako video. Izvi zvinotora 30 masekondi kusvika 2 maminitsi.

Your Talking-Head Video

Dhawunirodha MP4

Nezve SadTalker

SadTalker (CVPR 2023, Tencent ARC) i open-source kutaura-muromo mufananidzo kuti anobata mumwe muviri pfungwa kutaura chero audio. Zvichisiyana Wav2Lip variants, SadTalkerwo anobata muromo pose, blinks, uye pfungwa yezvakasikwa zvakawanda.

Kodhi uye zviyero zvinopihwa mvumo neMIT kubva pakutanga kusvika pakupera — hapana Llama, Gemma, kana chimiro chisingatengese — saka mavhidhiyo aunoita akasimba uye anobatsira kugoverwa kwekutengesa.

Mazano eBest Results

  • Usashandisa mafoto ane mhando yepamusoro, ane magetsi akanaka—maziso anoonekwa, muromo wakavharwa
  • Centered face, square kana 4:5 aspect ratio inoita zvakanaka
  • Kutaura kwakachena (sina mukurumbira) kunopa kunyatsoshanda kwe lips sync
  • Kubvumira GFPGAN yevatambi shots - doubles render nguva asi sharpens deta
  • Usashandisa chirongwa che Still kana uchida kutambanudza avatar yako

Lip Sync Video Plans

Kutanga zvakasununguka, kuvandudzwa kana iwe uchida zvakawanda

Free
  • 30-sekondi audio limit
  • 256 px output
  • "Still" preset chete
  • Hapana face enhancer
Inonyanya Kuzivikanwa
Free Account
  • 30-sekondi audio limit
  • Both "full" and "still" presets
  • 256 / 512 px output
  • GFPGAN face enhancer
Sign Up Free
Pro
  • 5-minute audio limit
  • GPU Priority Queue
  • API kuwanikwa (multipart kurodha)
  • Webhook kudzokorora
  • Kushandiswa kwekutengesa (MIT license)
Upgrade

Mibvunzo Inobvunzwa Kazhinji

Upload a photo chiso uye audio clip, uye AI inogadzira video yechiso achitaura audio nechokwadi lip mafambiro, head pose, uye blinks. Yakavakwa pa SadTalker (CVPR 2023), MIT-licensed kutaura-chiso model iyo animates chirevo pamwe nemouth shape.

Iyo yekupinda yemuviri inogona kuve JPG kana PNG vhidhiyo (kusvika 10 MB) kana yakafupi MP4 / WebM yekufambisa video (tinoshandisa yekutanga frame). Kufamba kweaudio inogona kuve MP3, WAV, M4A, kana FLAC kusvika ku10 MB.

Free accounts: kusvika 30 masekondi pa clip. Paying vashandisi: kusvika 5 maminitsi pa request. Longer audio zvinoreva refu render nguva uye yepamusoro character mutengo.

Lip sync video inoshandisa 1,000 characters pasecond yevhidhiyo yakagadzirwa. A 30-second clip = 30,000 characters. The cost is billed up front from your character balance and refunded automatically if generation fails.

Yeah — SadTalker code uye zvikamu zvemifananidzo zvinopihwa pasi peMIT license kubva pakutanga kusvika pakupera (sina Llama, Gemma, kana chero chinhu chisina kutengeswa). Mavhidhiyo aunoita ndeako chete uye unogona kushandisa zvekutengesa. Unofanira kunge uine kodzero dzemufananidzo wechiso uye zvemufananidzo wezwi zvaunotumira.

Pamusoro pe 30 masekondi e5-second clip paA100 server yedu, ichiwedzera zvakaenzana nenguva yevhidhiyo. Kubvumira GFPGAN face enhancer kunowedzera nguva yekudzokorora asi kunogadzira yakajeka, yepamusoro-mhando output.

Full preset (default) inobata pfungwa dzemuromo, kutambanuka kwemaso, uye kuratidzwa pamwe nemeso, zvichipa vhidhiyo ine pfungwa dzemuromo dzinotaura zvakajeka. Still preset inobata pfungwa dzemuromo uye inobata pfungwa dzemuromo chete — zvinokosha kana iwe uchida avatar shot ine pfungwa dzakachengeteka.

GFPGAN imhando yekugadzirisa maziso iyo inowedzera kunaka kwemashoko emaso mushure mekuisa maziso pamwe chete. Inoita kuti 256-pixel output ionekwe seyakaenzana ne512. Inoita kuti nguva yekuita iite seyakapetwa kaviri asi inoita kuti zviite sezvakakodzera kune zvifananidzo zvevatambi.

SadTalker inoratidzwa ne 256 px zvakaipa. Dzvanya pa 512 px saizi yekuona zviri nani (kuoma, yakakwira VRAM) kana shandisa GFPGAN enhancer kuti uone zvinyorwa zvemaziso. Kuti uwane mhedzisiro yakanaka, enda ku high-quality, well-lit portrait photo.

Yeah. Upload a MP4 kana WebM sechiso mudziyo uye isu tichashandisa yekutanga frame sechiratidzo kufambisa. For full video re-dubbing (per-frame muromo kuchinja), ona kuuya Dubbing Studio video pipeline.

Yeah. POST a multipart request to /api/v1/lipsync/ with face and audio fields, then poll /api/v1/lipsync/result/?uuid= until status is "completed". The response contains a URL to the rendered MP4. API access requires a paid plan.

SadTalker inoshandisa kusangana kwemuviri kuongorora uye kuchera chiso chinozivikanwa. Kuti uwane zvibodzwa, enda ku portrait ine munhu mumwe chete akatarisana, nemaziso anoratidza, uye nechimwe chinhu chisingaonekwe.
5.0/5 (1)

Chii chingatibatsira kuti tiite zvakanaka? Ruzivo rwako runogona kutibatsira kugadzirisa matambudziko.

Wagadzirira kutanga?

Sign up for free and get 15,000 characters. No credit card required.

Sign Up Free Ona Kubhadhara