ttlequals0/MinusPod
A self-hosted server that removes ads from podcasts using speech-to-text transcription and LLM-based ad detection.
Collecting fresh signals — velocity needs a few days of history.
collecting data…
star history
MinusPod transcribes podcast episodes locally with faster-whisper or NVIDIA Parakeet, or through remote APIs, then uses an LLM to identify and cut ad segments, with a second verification pass and an optional human review queue. It supports multiple LLM providers including Claude, Ollama, OpenRouter, and OpenAI-compatible endpoints. User corrections are turned into cross-episode ad patterns so repeat sponsors are detected without additional LLM calls. A web dashboard provides processing controls and a waveform editor for adjusting low-confidence cuts.
Frequently asked
- What is ttlequals0/MinusPod?
- A self-hosted server that removes ads from podcasts using speech-to-text transcription and LLM-based ad detection.
- Is MinusPod open source?
- Yes — ttlequals0/MinusPod is open source, released under the MIT license.
- What language is MinusPod written in?
- ttlequals0/MinusPod is primarily written in Python.
- How popular is MinusPod?
- ttlequals0/MinusPod has 503 stars on GitHub.
- Where can I find MinusPod?
- ttlequals0/MinusPod is on GitHub at https://github.com/ttlequals0/MinusPod.