A.I.Warper (@AIWarper)

LTX 2.3이 오디오와 시작 프레임만으로도 자연스러운 영상 생성이 가능하다고 언급됐다. 후처리 한 번을 더하면 품질이 좋아지고, 여러 개를 연결하면 저비용 콘텐츠 제작 에이전트로 활용할 수 있다는 점이 흥미롭다.

https://x.com/AIWarper/status/2034314551324463463

#ltx #aivideo #audiotovideo #contentcreation

A.I.Warper (@AIWarper) on X

I reiterate that LTX 2.3 can do some magic with just audio + start frame. A second pass to clean this up would work really well. Now imagine chaining a bunch of these together... easy content creation agent for free

X (formerly Twitter)

New breakthroughs unveiled at Davos push AI video to a new level: audio-to-video generation, agentic skill control, and vision-language models now open-source via Vercel’s LTX toolkit. Discover how consistency and customization are finally within reach. #AIVideo #AudioToVideo #OpenSourceAI #Davos

🔗 https://aidailypost.com/news/ai-video-hits-high-bar-new-tools-consistency-customization-davos

Angry Tom (@AngryTomtweets)

오디오를 업로드하거나 음성을 녹음하거나 LTX에서 소리를 생성하면 이를 자동으로 비디오로 변환해 보여주는 'Audio to Video' 기능을 소개하는 트윗. 오디오 → 비디오의 멀티모달 변환 워크플로우를 간단히 안내함.

https://x.com/AngryTomtweets/status/2014134053348262065

#audiotovideo #multimodal #ltx #ai

Angry Tom (@AngryTomtweets) on X

This is Audio to Video. Upload audio, record your voice, or generate sound in LTX, then watch it transform into video.

X (formerly Twitter)

Yoav HaCohen (@yoavhacohen)

ElevenLabs와 협력해 LTX Studio의 Audio-to-Video 모델을 ElevenLabs Creative Platform에서 출시한다고 발표했습니다. 오디오를 한 번의 흐름으로 비디오로 전환하는 멀티모달 생성 기능 도입으로 크리에이티브 플랫폼의 생성형 미디어 워크플로우가 강화될 것으로 보입니다.

https://x.com/yoavhacohen/status/2013873219023888633

#audiotovideo #multimodal #elevenlabs #ltxstudio #generativeai

Yoav HaCohen (@yoavhacohen) on X

Audio → video in one flow. Thrilled to team up with @elevenlabsio to launch @LTXStudio’s Audio-to-Video model on the ElevenLabs Creative Platform.

X (formerly Twitter)

Justine Moore (@venturetwins)

새로운 Audio-to-Video 기능은 사용자가 오디오를 생성하거나 업로드하면 이를 기반으로 영상을 만들어 줍니다. 오디오(생성형)를 사용할 경우 음성 합성은 ElevenLabs의 스피치 모델을 사용해 다양한 목소리 선택과 감정·톤 제어가 가능하다는 점을 강조합니다.

https://x.com/venturetwins/status/2013687988673262017

#audiotovideo #elevenlabs #speechsynthesis #voice #multimodal

Justine Moore (@venturetwins) on X

@LTXStudio This is the new Audio-to-Video feature. If you decide to generate audio (like I did) - the speech model behind this is @elevenlabsio. That means there's a bunch of great voices to choose from AND you can control emotion and tone. Here's a quick demo of how it works ⬇️

X (formerly Twitter)

Justine Moore (@venturetwins)

LTXStudio가 오디오를 업로드하거나 생성하면 해당 오디오에 맞춰 립싱크된 영상을 생성하는 audio-conditioned video 모델을 공개했습니다. 캐릭터 보이스의 일관성 유지와 AI 인플루언서, 캐릭터 기반 콘텐츠 제작에 큰 영향을 줄 수 있는 멀티모달 영상 생성 기능입니다.

https://x.com/venturetwins/status/2013686150494593466

#audiotovideo #ltxstudio #lipsync #multimodal #aivideo

Justine Moore (@venturetwins) on X

We've got a new audio-conditioned video model 👀 @LTXStudio now lets you upload or generate audio, and then creates a lip-synced video. This is going to be HUGE for consistent character voices + AI influencers. I tested it w/ an animation of my pets. More on how to use it 👇

X (formerly Twitter)
Kargo unveils Narrative: Audio-to-Video format expands CTV advertising access: New Narrative product transforms audio ads into engaging CTV content, opening opportunities for diverse advertisers and enhancing audience reach. https://ppc.land/kargo-unveils-narrative-audio-to-video-format-expands-ctv-advertising-access/?utm_source=dlvr.it&utm_medium=mastodon #CTVAdvertising #AudioToVideo #DigitalMarketing #AdvertisingInnovation #Kargo
Kargo unveils Narrative: Audio-to-Video format expands CTV advertising access

New Narrative product transforms audio ads into engaging CTV content, opening opportunities for diverse advertisers and enhancing audience reach.

PPC Land

oh nice.

here's another trick with audio and ffmpeg:

IF you have an mp3/ogg/flac/opus file that has an EMBEDED COVER in ID3 tag, ffmpeg will be very smart with it - all i did here was:

ffmpeg -i levitation.flac -pix_fmt yuv420p levitation.mp4

if ffmpeg complains, try setting image size with -s parameter:

ffmpeg -i input.flac -pix_fmt yuv420p -s 640x640 output.mp4

#ffmpeg #audiotovideo #tricks #mastodon https://toot.si/media/sCxOCF92_uDFii4tQY0