HN Showcase
Back to browse
GitHub Repository

Open-weight only, video-aware podcast dubbing with ASR alignment, diarization, LLM translation, and voice-cloned TTS. Local inference.

2 starsPython

Dub Any Podcast in Your Language Using Local Models

by ggamecrazy·Jul 26, 2026·1 point·0 comments
Visit ProjectView on HN

AI Analysis

●●SolidShip ItNiche Gem

Local podcast dubbing pipeline with speaker diarization, though translation quality still needs big models.

Strengths
  • •End-to-end pipeline handles diarization, translation, and voice cloning in a single automated flow.
  • •Resumable job artifacts allow restarting failed stages without reprocessing the entire video.
  • •Deep-ducking original audio underneath the dub preserves context better than full replacement.
Weaknesses
  • •Translation quality degrades significantly without using large, non-local proprietary models.
  • •Requires 16GB+ RAM on M-series Macs, limiting accessibility for older hardware users.
Category
AI/ML
Target Audience

Content creators and podcasters needing localization

Similar To

Rask.ai · ElevenLabs Dubbing · HeyGen

Post Description

Worth mentioning that the pipeline uses all open-weight models, though for the demo I used Kimi K3 for the translation piece due to the Kimi founder being interviewed.

Have a listen for yourself: https://www.youtube.com/watch?v=92BQg2oozBg

The pipeline can be mostly run locally on an M-series Mac with at least 16 GB of RAM.

The translation stage is the main quality constraint but local models are getting better and better. Could see a specialized 30B model be more than enough here.

Did this on a whim so I could listen to the Kimi founder being interviewed. What stood out to me is how fast the software and even the models themselves seem to be getting commoditized. Far faster than I ever anticipated.

Built using Kimi Code, thought the model needed some guidance, so not a 1-shot effort quite yet.

Similar Projects