konstantinpaulus's banner

konstantinpaulus

@konstipaulus • 5,310 subscribers

CEO @diffusionhq (YC F24) | Github: k9p5

Shorts

Everyone’s doing this trendy hero effect with a plain <video> tag. I tried it with WebCodecs for extra performance. Repo + tutorial below

Everyone’s doing this trendy hero effect with a plain <video> tag. I tried it with WebCodecs for extra performance. Repo + tutorial below

101,208 Aufrufe

I finally landed on a memory-optimal way to render audio waveforms in the browser. The full audio is decoded once, then chunked into 2s bins. Each bin stores peaks at 800 peaks/sec in a Uint8Array. Bins are persisted individually in IndexedDB, which keeps the data as raw Uint8 arrays and is faster to read than OPFS. Local or session storage wouldn’t work here. A 2h video ends up as ~5.76 MB cached on disk. While scrolling or zooming the timeline, I load only the visible bins and downsample them to the required resolution. Everything runs async and I never keep more than a single bin in memory at once, ~800 bytes. Rendering supports mixed resolutions in the same pass since bins arrive async. By caching a read index and sorting bins ascending, each peak lookup stays O(1). On refresh, I wipe the IndexedDB so cache size never gets out of hand.

I finally landed on a memory-optimal way to render audio waveforms in the browser. The full audio is decoded once, then chunked into 2s bins. Each bin stores peaks at 800 peaks/sec in a Uint8Array. Bins are persisted individually in IndexedDB, which keeps the data as raw Uint8 arrays and is faster to read than OPFS. Local or session storage wouldn’t work here. A 2h video ends up as ~5.76 MB cached on disk. While scrolling or zooming the timeline, I load only the visible bins and downsample them to the required resolution. Everything runs async and I never keep more than a single bin in memory at once, ~800 bytes. Rendering supports mixed resolutions in the same pass since bins arrive async. By caching a read index and sorting bins ascending, each peak lookup stays O(1). On refresh, I wipe the IndexedDB so cache size never gets out of hand.

32,259 Aufrufe

Videos

Anya Rossi

sweetdream.ai

SweetDream.ai•Sponsored•Livecam

Watch Anya Live

Anya is streaming live right now! Join her private show and enjoy exclusive content.

Exclusive private shows

1.2k viewers online

Private Show

Join now for exclusive access

Free preview available • Premium content

video editing is stuck in 2005, time for something new introducing diffusion the first infinite canvas for video and motion graphics like figma, but for editing

video editing is stuck in 2005, time for something new introducing diffusion the first infinite canvas for video and motion graphics like figma, but for editing

konstantinpaulus

128,238 Aufrufe • vor 1 Monat

Introducing Diffusion Studio Pro. The world’s first multimodal AI studio for professionals. Think CapCut + Cursor. Local first, browser based editor (non linear). Free unlimited tier. Fully agentic sidebar with access to 16+ generative models. Comment “DSPRO” to get early access!

Introducing Diffusion Studio Pro. The world’s first multimodal AI studio for professionals. Think CapCut + Cursor. Local first, browser based editor (non linear). Free unlimited tier. Fully agentic sidebar with access to 16+ generative models. Comment “DSPRO” to get early access!

konstantinpaulus

348,156 Aufrufe • vor 1 Jahr

Free professional-grade video editing, in your browser. No login required. Introducing Diffusion Studio Pro: The internet’s first and best browser-native video editor. Link Below - Comment "DSPRO" to get the JSON prompt template.

Free professional-grade video editing, in your browser. No login required. Introducing Diffusion Studio Pro: The internet’s first and best browser-native video editor. Link Below - Comment "DSPRO" to get the JSON prompt template.

konstantinpaulus

89,164 Aufrufe • vor 8 Monaten

WebGPU inference might be the most underrated tech in AI right now. This entire audiobook was generated in the browser on my M1 MacBook Pro. - Kokoro 82M for TTS - Whisper Base for transcription/captions It runs so fast with WebGPU that most people won't even realize it's happening on their own hardware.

WebGPU inference might be the most underrated tech in AI right now. This entire audiobook was generated in the browser on my M1 MacBook Pro. - Kokoro 82M for TTS - Whisper Base for transcription/captions It runs so fast with WebGPU that most people won't even realize it's happening on their own hardware.

konstantinpaulus

62,967 Aufrufe • vor 9 Monaten

The world’s most intuitive AI canvas is here. We built an agentic canvas for generating pro-level creative assets. Think Figma meets Sora. Comment “Canvas” and we’ll DM you an invite code.

The world’s most intuitive AI canvas is here. We built an agentic canvas for generating pro-level creative assets. Think Figma meets Sora. Comment “Canvas” and we’ll DM you an invite code.

konstantinpaulus

13,538 Aufrufe • vor 11 Monaten

Keine weiteren Inhalte verfügbar