Voquill
The open source alternative to WisprFlow
Start new thread

Voquill - The open source WisprFlow alternative

by•
Voquill is the open source alternative to WisprFlow. Type 4x faster by using your voice. It works in any app and runs natively on any operating system (MacOS, Windows, or Linux). Whether you're using agent mode or AI dictation mode, Voquill understands your intent and turns what you say into beautifully formatted text.

Add a comment

Replies

Best
Maker
📌
Hey everyone! 👋 I'm Josiah, one of the co-founders of Voquill. We're building this product because we found AI dictation to be an incredibly useful tool. We believe that great tools should be transparent and available for everyone to contribute to, use, and have full confidence that their data is safe. That's why we're building Voquill as a fully open-source alternative to WisprFlow. Voquill works in any app and runs on any operating system. You can run it using our cloud, connect it to an API of your choosing, or even run it entirely locally. For enterprise users, we can also set up a private cloud so you can use Voquill with 100% confidence that your data stays completely private. Would love to hear what you think! Thanks, Josiah, Henry, & Michael
Sounds great. Congrats on the launch . Am I right in thinking that this is only available on Desktop and chrome browser on mobile at the moment? If so, when do you think this will be available to use on mobile in Safari ?
Maker

Thanks Simona! We’re sunsetting our Chrome extension in favor of this new Desktop app since it's more versatile. We'll be starting work on our mobile version next. Hoping to have it available in the next few weeks.

When someone is choosing between Voquill and a polished closed-source option like Wispr Flow, what are the differences that actually matter in daily use?
Maker

 Great questions. For me it’s transparency and privacy. Voquill’s code is open, so you can trust your data isn’t being harvested or trained on. And for enterprises, you can run dictation entirely on-prem. This is super important for organizations that are working with private data that shouldn’t leave their network. On top of this, you get the added benefit of owning what you use; if you don’t like something, you can fix it yourself

Interesting. Can I “consult” with the AI? That is, I say who I’m writing to, what I want to write about, and what points to emphasize, and the AI generates the text itself?

Maker

 Yes, exactly! Bearing in mind that AI mode is still in open beta, what you described there is what we’re working towards. It’s been a super handy feature so far

Maker

 This is rad, I’m definitely adding “I use arch btw” onto the downloads screen lol

what does latency look like if we run it locally on a M4 macbook pro?
Maker

 It depends on how long your transcription is. For a normal sized transcript, it’ll take a couple seconds. Running with an NVIDIA GPU or against the cloud is crazy fast.

The cross-platform support and local model option are attractive for developer workflows. I'm curious about language handling—does the speech recognition support multiple languages like Japanese, and if so, does the AI formatting adapt to each language's writing conventions?

Maker

 Great questions! Voquill supports 100+ languages (including Japanese). The AI adapts to the language you're currently using and formats it appropriately for that language.

Can we bring our own fine-tuned models if we run it locally?

Maker

 Absolutely! Voquill can work with Ollama, so you can bring your own LLMs. For transcription, we'd need a standard interface to do so, but can definitely support that. Feel free to reach out on Discord to discuss further

Does Voquill support multi-language dictation?

Maker

 It does indeed! Voquill supports 100+ languages, thanks to the Whisper model it uses under the hood.

The opensource is the magic!

Maker

 amen to that!