Upload Audio

Add new assets to your radio station library

Podcasts will be transcribed and indexed for AI-powered semantic search.

Pipeline 2 runs the existing Whisper search processing and Gemini Studio script creation in parallel.

On — this upload must complete both the Whisper search and Gemini Studio branches.

Drop audio files here

or click to browse

Supports MP3, WAV, M4A, FLAC, OGG, AAC

What happens after upload?

  1. Audio is transcribed using Whisper AI
  2. Transcript is split into semantic chunks
  3. Each chunk is converted to embeddings
  4. Embeddings are stored for fast similarity search

Processing typically takes 1-2 minutes per 10 minutes of audio.