Ito is an open source voice assistant for Mac and Windows that transforms your intent into smart text in any app. Speak naturally to write emails, messages, or code without typing. Say intent, not just words.
@nasira_bibi , great question! I'm one of the engineers working on Ito. Many of the speech-to-text products out there ask for fairly extensive permissions, but they are closed-source. We saw the value of being able to dictate easily across all apps, but we also felt uneasy about not knowing what was happening under the hood, which is why Ito is open-source.
As far as our roadmap, we keep adding ideas and taking feature requests from our community. We have our own ideas of how to push the product too, since all of us use it everyday across engineering, product, and marketing.
Personally, I am most excited about our future work on intent-dictation, because that will allow Ito to write what I mean instead of what I say.
@sentry_co Great question! We're open source, so you can audit exactly what is happening with your voice data, examine how the whole app works, and verify how the permissions granted to Ito are used. Our long-term roadmap includes enabling a plugin ecosystem so people can add custom-tailored integrations and specific functionality for whatever apps they want to use Ito in. We also have an expanding list of advanced settings that power users can tweak to really push extra productivity out of their own voice assistant.
@mike_ito I remember Wispr got some heat around data privacy yes. And blamed it on electron and moving fast. Fair enough. The plugin system sounds interesting. What kind of plugins?
@sentry_co Plugins would be application specific. They would enable gathering additional context, additional formatting rules, and further in the future, actions. If you imagine a plugin for cursor for example, there are many different text inputs that would change how you want Ito to respond. If it's in the AI panel, normal formatting is probably okay but in the terminal panel, you want valid commands. We're building out some of our application specific handlers now and we want to generalize this for external contributors.
@mike_ito I C. So Sort of filters. Filtering input and output. Sort of like an interpreter with AI. That could make speech to text really interesting! Like autocomplete tabbing but through speech.
Report
As a non-native English user, what I care about most is whether it can reliably handle accents, mixed Chinese-English speech, and proprietary terms. I hope the experience can be improved in this area.
Congrats Barron and team. VibeTyping feels like the missing link between dictation and a real assistant. Intent over words, open source, and a hotkey that works anywhere is a killer combo. How do custom vocab and style presets work?
@alexcloudstar There's a dictionary tab where you can add any custom vocabulary. And in the advanced settings, you can change the prompts that are used.
Report
How did you get around the native Mac transcription being terrible and whisper being slow? Did you take the approach most do where you approximate with dictation first and then hot swap text a few seconds later after it’s processed through a better model?
Wispr Flow
Spokenly
Solid product how did you guys come up with the ideas?
Ito
@nasira_bibi , great question! I'm one of the engineers working on Ito. Many of the speech-to-text products out there ask for fairly extensive permissions, but they are closed-source. We saw the value of being able to dictate easily across all apps, but we also felt uneasy about not knowing what was happening under the hood, which is why Ito is open-source.
As far as our roadmap, we keep adding ideas and taking feature requests from our community. We have our own ideas of how to push the product too, since all of us use it everyday across engineering, product, and marketing.
Personally, I am most excited about our future work on intent-dictation, because that will allow Ito to write what I mean instead of what I say.
DiffSense
Talking product position. How are you positioning your self differently than wispr?
Ito
@sentry_co Great question! We're open source, so you can audit exactly what is happening with your voice data, examine how the whole app works, and verify how the permissions granted to Ito are used. Our long-term roadmap includes enabling a plugin ecosystem so people can add custom-tailored integrations and specific functionality for whatever apps they want to use Ito in. We also have an expanding list of advanced settings that power users can tweak to really push extra productivity out of their own voice assistant.
DiffSense
@mike_ito I remember Wispr got some heat around data privacy yes. And blamed it on electron and moving fast. Fair enough. The plugin system sounds interesting. What kind of plugins?
Ito
@sentry_co Plugins would be application specific. They would enable gathering additional context, additional formatting rules, and further in the future, actions. If you imagine a plugin for cursor for example, there are many different text inputs that would change how you want Ito to respond. If it's in the AI panel, normal formatting is probably okay but in the terminal panel, you want valid commands. We're building out some of our application specific handlers now and we want to generalize this for external contributors.
DiffSense
@mike_ito I C. So Sort of filters. Filtering input and output. Sort of like an interpreter with AI. That could make speech to text really interesting! Like autocomplete tabbing but through speech.
As a non-native English user, what I care about most is whether it can reliably handle accents, mixed Chinese-English speech, and proprietary terms. I hope the experience can be improved in this area.
minimalist phone: creating folders
Where can I find the list of supported languages?
Ito
@busmark_w_nika Thanks for checking us out! Here's the full list of languages we currently support:
English, Chinese, German, Spanish, Russian, Korean, French, Japanese, Portuguese, Turkish, Polish, Catalan, Dutch, Arabic, Swedish, Italian, Indonesian, Hindi, Finnish, Vietnamese, Hebrew, Ukrainian, Greek, Malay, Czech, Romanian, Danish, Hungarian, Tamil, Norwegian, Thai, Urdu, Croatian, Bulgarian, Lithuanian, Latin, Māori, Malayalam, Welsh, Slovak, Telugu, Persian, Latvian, Bengali, Serbian, Azerbaijani, Slovenian, Kannada, Estonian, Macedonian, Breton, Basque, Icelandic, Armenian, Nepali, Mongolian, Bosnian, Kazakh, Albanian, Swahili, Galician, Marathi, Punjabi, Sinhala, Khmer, Shona, Yoruba, Somali, Afrikaans, Occitan, Georgian, Belarusian, Tajik, Sindhi, Gujarati, Amharic, Yiddish, Lao, Uzbek, Faroese, Haitian Creole, Pashto, Turkmen, Nynorsk, Maltese, Sanskrit, Luxembourgish, Myanmar (Burmese), Tibetan, Tagalog, Malagasy, Assamese, Tatar, Hawaiian, Lingala, Hausa, Bashkir, Javanese, Sundanese, and Cantonese.
minimalist phone: creating folders
@julian_gomez11 aaa, cool, my is included! Thank you! :D
Makers Page
Congrats Barron and team. VibeTyping feels like the missing link between dictation and a real assistant. Intent over words, open source, and a hotkey that works anywhere is a killer combo. How do custom vocab and style presets work?
Ito
@alexcloudstar There's a dictionary tab where you can add any custom vocabulary. And in the advanced settings, you can change the prompts that are used.
How did you get around the native Mac transcription being terrible and whisper being slow? Did you take the approach most do where you approximate with dictation first and then hot swap text a few seconds later after it’s processed through a better model?
Lancepilot
Simple yet powerful. Loved the idea ( :
Ito
Thanks so much for giving Ito a try! @istiakahmad