----------------🛠️ Tool===================MinusPod: self-hosted podcast ad removal with local transcription and LLM dete...

----------------🛠️ Tool===================MinusPod: self-hosted podcast ad removal with local transcription and LLM detectionMinusPod is a self-hosted server that re-cuts podcast RSS feeds to strip ad segments before they ever reach the player. The pipeline transcribes episodes locally with faster-whisper or NVIDIA Parakeet (GPU or CPU), runs an LLM over the transcript to locate ad segments, verifies each cut in a second pass, and republishes the feed with versioned audio. The LLM layer is bring-your-own: Anthropic, OpenRouter, Ollama, System One, or any OpenAI-compatible endpoint, switchable at runtime with per-stage models and editable prompts.Key features• Two-stage detection. An LLM finds ads in the transcript, a second pass verifies the cut, and low-confidence boundaries wait in a review queue with a waveform editor. An optional reviewer confirms or adjusts each boundary.• Audio evidence. The transcript is treated as one signal among several: loudness and DAI transitions, a cross-...

Read Original

Related

Mastodon discussion 28m ago

“Higgsfield VS Adobe:AIクリエイティブ市場の覇権争い――生成AIは巨人の堀を崩せるか #生成AI #動画生成 #Adobe #Higgsfield” (1 user) https://htn.to/3uQXpz71TY...

“Higgsfield VS Adobe:AIクリエイティブ市場の覇権争い――生成AIは巨人の堀を崩せるか #生成AI #動画生成 #Adobe #Higgsfield” (1 user) https://htn.to/3uQXpz71TY #ソフトウェア #情報産業 #コンピュータグラフィ #経営戦略 #マーケティング #映像制作 #人工知能 #ai