RiverScript is an AI-powered transcription app for Windows and macOS built around one idea: if you can hear it on your computer, RiverScript can turn it into text.

Live Recording Transcription captures system audio directly - meetings, podcasts, lectures, any video - no matter which app it's playing in. It doesn't route around your microphone or hope for the best; it grabs the actual audio your computer is playing and transcribes it live. Your audio, your rules.

Under the hood, RiverScript runs a multi-model AI architecture combining leading speech recognition models from ElevenLabs, OpenAI, and Deepgram, so accuracy holds up across accents, background noise, and multiple speakers.

Built for real people - students, content creators, journalists, lawyers, healthcare workers - anyone who needs spoken audio turned into usable text without babysitting the process.

Features:

- Upload audio or video files up to 50GB or 8 hours long, including batch uploads

- Live Recording Transcription of system audio on Windows and macOS

- Interactive editor with an inline media player

- Speaker diarization with color-coded labels

- Clickable, timestamped subtitles

- AI translation into multiple languages

- AI-generated summaries

- Ask AI anything about your transcript

The desktop client is native and built on Rust for speed, not wrapped in a browser. Born out of a real need for a transcription tool that just works, not a marketing pitch - free to start, no credit card required.

RiverScript: transcribe everything.