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

AI Audio Inpainting

Replace a section of audio with AI-synthesized speech that matches the surrounding voice. Fix a bad take without re-recording the whole thing.

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

Kuisa Audio kuInpaint

500 characters per second of audio replaced

Drag & drop your file here, or browse

Supports MP3, WAV, FLAC, OGG, M4A. Max 500 MB (2 GB on paid plans). Up to 10 minutes.

file.mp3

0 MB

Izvozvo zvemavhidhiyo — bvisa kuti uwane iyo yakaipa vhidhiyo

0.00s / 0.00s

Kugadzirisa

0 / 500 mavara
80ms ndiyo default — match-kubvisa vanonzwa zvakasikwa, hapana anonzwa double-trigger.
Sign up for free to use audio inpainting

Kuita vhidhiyo...

Slicing → cloning surrounding voice → splicing with crossfade
Kutora nguva? Your result ichaonekwa mu generation history kana wagadzirira.
Inpainted Audio Ready

Pashure (Original)

After (Inpainted)

Dhawunirodha Inpainted Audio

Maitiro ekushandisa Audio Inpainting

Inpainting icho ndechekunyora kwezwi kunoita sekuti ndicho chigadzirwa chePhotoshop chekunyora kwezwi. Tinotora mashoko kubva mumashoko ari kutenderera pane zvawakasarudza, tinogadzira mashoko matsva mumashoko, uye tinozvibatanidzazve ne crossfade.

Best results: bvisa panenge 3 masekondi emashoko akachena pamberi pekuchinja point kuti cloner iwane zvinhu zvakanaka zvekuita reference.

Tips for Best Results

  • Tenga zvakanyorwa pakati seyakaomarara sezvinobvira - chete zvakaipa kutora
  • Tevere, shanduro itsva inofanira kunge iri nehurefu hwakafanana nehurefu hwechinyorwa icho ichishandura.
  • Sarudza rurimi kuti rifananidze nezwi remutauro wekutanga kuti rifananidze zvakanaka
  • 80ms crossfade ndeyezvinowanzoitika zvisingaratidzike; bump kune 150ms kana iwe uchinzwa kutsvedza
  • Kuti uite shanduko dzenguva refu (>10s), funga nezvekunyorazve zvese zvinotaurwa.

Maitiro eAI Audio Inpainting Works

Surgical edits, voice-matched, pasina re-recording session.

Step 1

Kuisa

Kuisa mashoko ako uye shandisa mutsara wezvinyorwa kuti uone kutangira/kupera kwechikamu chaunoda kuchinja. Dzvanya pane mutsara waunoda kuchinja.

Step 2

Voice Clone + Synthesize

Tichatora kusvika pa12 masekondi ezvokutaura zvakachena zvinotarisana nezvokutaura zvawakasarudza, tichagadzira vhidhiyo yemutauro wemutaura, uye tichagadzira mashoko matsva mumashoko aya.

Step 3

Crossfade Splice

Iyo synthesized clip inoiswa muoriginal recording ne crossfade ine simba rimwe chete pazvimwe zvidimbu zve edit. Mativi haaratidzike.

Audio Inpainting Plans

Kutanga zvakasununguka, kuvandudzwa kana iwe uchida zvakawanda

Free
  • Kusvika 10-minute source faira
  • 500-character replacement text
  • 4-second inpaint per request
  • 80ms crossfade splice
  • OpenVoice + CosyVoice 2 backends
Inonyanya Kuzivikanwa
Free Account
  • Kusvika 10-minute source faira
  • 5,000-character replacement text
  • Tunable crossfade (0-250ms)
  • Kubvisa-mufananidzo-we-mutauro
  • Kuumbwa kwebhuku rezita + re-edit
Sign Up Free
Pro
  • Kusvika 30-minute source faira
  • 100,000-character replacement text
  • GPU Priority Queue
  • API kuwanikwa (/v1/audio-inpaint/)
  • Kuisa mutsara
Upgrade

Mibvunzo Inobvunzwa Kazhinji

Audio inpainting (akanziwo audio fill kana kutaura overdub) inokutendera kuti ushandure chikamu cheiyo yazvino audio recording nenyaya nyowani yeAI-synthesized speech iyo inofananidza nezwi rekutanga. Iyi ndiyo audio equivalent yePhotoshop's content-aware fill - peani pamusoro pechikamu chauchada, tinya izvo zvaunofanira kunge uchiona, uye AI inogadzira kuchinja kwakasununguka.

Kutaura kweAI kunoita kuti zvive nyore kushandura mashoko kubva pafoni yako kuenda pakombuta yako. Nekudaro, kana iwe uchida kushandura mashoko kubva pafoni yako kuenda pakombuta yako, unogona kuisa mashoko kubva pafoni yako kuenda pakombuta yako. Nekudaro, kana iwe uchida kushandura mashoko kubva pafoni yako kuenda pakombuta yako, unogona kuisa mashoko kubva pakombuta yako kuenda pakombuta yako.

Usaishandisa kana iwe ukawana rimwe chete rimwe zita rakanaka, zita rakaipa, zita rakaipa, zita rakaipa, kana nyaya yakaipa muzita rakanaka. Kunyorazve mashoko ose kunokonzera kuti mashoko asingafananidzike nevamwe vemubasa. Kunyorazve kunogadzirisa chete zvaunoda kugadzirisa, asiwo kuchengeta mamwe mazwi akachengeteka.

Vashandisi vemahara vanogona kuisa mafaera kusvika 10 maminitsi. Vashandisi vemahara vanogona kuisa mafaera kusvika 30 maminitsi. Tenzi wekugadzirisa wezvinyorwa zvachose ane huwandu hwemavara 500 evashandisi vemahara, 5,000 evashandisi vemahara, uye 100,000 evashandisi vemahara.

12 sekondi yezwi rinotarisana nechirongwa sechiratidzo chekutaura, izvo zvinokwana kuti chero imwe yemhando dzedu dzinokwanisa kushandura (OpenVoice, CosyVoice 2) ione timbre, pitch, uye nzira yekutaura yemutaura. Kuti uwane mhedzisiro yakanaka, bvisa 3 sekondi dzemashoko akachena pamberi pechirongwa.

Tichashandisa 80ms equal-power crossfade pazvimwe zviratidzo zve splice (head→replacement uye replacement→tail) sezvatinoda. Unogona kudzora izvi kubva pa0ms (hard cut) kusvika pa250ms kuburikidza neCrossfade slider. Ma crossfades akareba anoficha kuchinja zvakanyanya asi anogona kutambanudza zvakapetwa-zvinyorwa panzvimbo yemuganhu.

Kuisa mashoko mumitauro inoteverwa neimwe nzira imwecheteyo yekuita mashoko. Tinoshandisa OpenVoice kune dzimwe nyika uye CosyVoice 2 kune Chinese, Japanese, neKorean. Unogona kuchinja iyo model mune advanced settings.

Iwe unobhadharwa 500 characters pasecond yeaudio yakagadziridzwa. 4-second fix inodhura 2,000 characters. Iyo mutengo inoenderana nenguva yekushandura tenzi, sezvo iyo inotungamira clone synthesis inodzorwa nenguva yekushanda yeiyo nyowani clip, kwete nenguva yekunyora.

Kunyangwe iwe uchida kushandura kana kushandura, unogona kushandura kana kushandura chero chinhu chaunofarira. Kushandura kana kushandura chero chinhu chinokanganisa hunhu hwedu, kana kushandura kana kushandura chero chinhu chisina kunaka, kana kushandura chero chinhu chisina kunaka, kana kushandura chero chinhu chisina kunaka.

Kukanda clip kunosiya chigaro chinoonekwa munguva nemweya; kutambanudza mapikicha maviri kunosiya kusingaenzaniswi kwechimiro. Kuisa mufananidzo muvhidhiyo kunozadza chigaro nemashoko anoenzanisa nezwi rinotarisa, kuitira kuti vaverengi vaone vhidhiyo isiri kumbomira, ine pfungwa yezvinyorwa.

Yes — POST to /v1/audio-inpaint/ with the audio file, start_sec, end_sec, and replacement_text. The endpoint returns a job UUID; poll /v1/speech/results/?uuid= to retrieve the inpainted audio when ready. See API docs for details.

ElevenLabs Speech-to-Speech inodzokorora zvese zvemutauro kubva pakutanga muzwi rakatarwa. Nhasi, kuisa mashoko mumifananidzo kunoitwa nechiremba: inogadzirisa chete iyo yakanyorwa nzvimbo, inochengeta chero imwe byte yemutauro wacho wakanyorwa, uye inoenzanisa iyo nyowani clip nezwi rakatarisana nebasa, kwete nebasa rakasiyana.
5.0/5 (1)

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

Fix Your Audio mumasekondi

Rekodha mashoko akanyorwa neAI uye shandura chero chikamu chekunyora neAI-yakagadzirwa mashoko anoenzana nezwi rayo rechinyorwa.

Sign Up Free Ona Kubhadhara