Launched this week
Whispering is an open-source, local-first transcription app. Use local and cloud models, chain custom transforms, and most importantly, keep your audio local on-device. Fast, ergonomic, and MIT-licensed. Let’s make closed-source apps obsolete. 🚀
Whispering
Hey everyone 👋 Braden here, creator of Whispering.
For years, I relied on transcription tools that were *almost* good, but they were all closed-source black boxes. Even those claiming to be “local” left me wondering where my audio really went. And when one pivoted or shut down, my workflow broke overnight.
So I built Whispering. It’s open-source, local-first, and most importantly, transparent with your data. Connect it to ultra-fast APIs like Groq Cloud, or run it entirely offline via Speaches. There is no middleman. For me, the features were good enough that I left my paid tools behind (I used Superwhisper and Wispr Flow before).
Whispering is transparent, extensible, and fully hackable under an MIT license. Chain custom transformations for grammar fixes, translation, or formatting. My favorite part? Hands-free, voice-activated dictation. I use it for several hours a day, from coding to thinking out loud while carrying pizza boxes back from the office.
Your data stays yours—always. We give you control over how and where it’s processed. Want the speed of the cloud? Connect directly to providers without a middleman. Prefer full privacy? Run everything offline. You can switch anytime, and you’ll always know where your audio goes and how it’s handled. We hope to add more offline models in the future.
This is the first app in the Epicenter ecosystem: a suite of interoperable, local-first tools with a shared, user-owned memory layer. We’re starting with transcription, but we’re not stopping here. Our eventual goal with Epicenter is to leverage that shared memory, and connect transcription to everything you think, write, and build, all within a trustworthy, local-first environment.
💻 Download Whispering: https://github.com/epicenter-so/...
⭐ Star on GitHub: https://github.com/epicenter-so/...
💬 Join the Discord: https://go.epicenter.so/discord
Would love your feedback, ideas, and roasts. Everything’s MIT licensed, so copy whatever you want. Let’s make closed-source productivity apps obsolete!
Autumn
Whispering is amazing! Can't wait for the devices to ship 🫶
Whispering
@ay_ush Not amazing as you and Autumn!
Is there any maximum limit of Audio ?
Whispering
@adarshreddy Yes, although it depends on provider! Most providers limit the audio size to 25 MB. Most recordings won't reach this size if they're only several minutes. You can also get around this by compressing the audio file with tools like ffmpeg, or splitting up the files and transcribing them in segments!