Speak, and polished text lands in any app. Record a call and get notes with who said what. Select anything to hear it in 950+ voices — free and fully on-device when you want.
Replies
Best
Really love the idea of keeping speech-to-text fully local — privacy-first tools like this are underrated. Does it handle accented English well?
Report
the local mode is the killer feature here — privacy-first dictation without needing an internet connection is huge for people working with sensitive docs. how does the accuracy compare between fast mode and local mode? and is the local model something like whisper under the hood, or a custom model?
@abhisheksinha123 So the accuracy of both the cloud mode and the local mode are nearly identical since they essentially use the same models. The difference is in the amount of AI polish that can be applied. With a cloud model, you can have a lot more degrees of quality versus on device; it's going to be pretty much exactly what you said. All the models we use locally are clearly labeled in your settings section and you can choose between them. And yes, Whisper and Parakeet are the 2 best models we have found.
Report
Local and free is a great combo. Most speech to text tools either need an internet connection or want a subscription for decent quality. How accurate is it compared to something like the built in Mac dictation?
@hiyamojo all of these tools basically use the same models. The difference might be in the instructions in the polish. We use all the top of the line models like Whisper from OpenAI or Parakeet from NVIDIA. Compared to Apple Dictation it 1000x better than that.
Report
a great free solution. I really like it. But it seems that 4000 words are not enough for me. I'll buy the Walkie Pro version
@cherryry1106 happy to offer you 3 month free. Just use the code THANKYOU3 at checkout.
Report
Thank you for offering this kind of local product! I love being able to use my own machine’s processing power to do these things without having to set up an entire development environment just for this.
As a Portuguese speaker, I wanted to ask you if it would be possible to “force” the adoption of a specific word in the dictation, since often, when using a multilingual model, it tries to interpret my Portuguese audio as English, which ends up causing some confusion. Would a small update be feasible that gives the option to “force” the recognition of a specific language?
@llanmelo Yes, if you set your primary language in the app as Portuguese, then it should be able to understand specifically what you're asking. We also do have some routing that happens behind the scenes, so it understands if you're speaking Portuguese or English or something else.
@conduit_design Thanks for sharing but I am still not following your feedback. Do you mind elaborating more. What exactly breaks about the voice when you talk alot. If you can provide a concrete example we can troubleshoot the issue. It could be a number of things including your machine specs, having a certain feature toggled on, running too many programs at the same time, etc.
@adam_perlis No, this is a human thing not a maschine thing. So a human can speak about 10k words before their voice gets brittle, our vocal chords are vibrating at very high freequencies etc. A very active day of coding often involves writing 10k words to an AI. Thats the tension. You can often see it in vibe code videos of devs that talk to their AI. their voice is gone. Esp with people that dont talk alot normally. They have way less than 10k spoken words a day. Maybe no more than 1k words. But there are things you can do. like 10min rest every 50 min. Also maybe analysying brittle ness with AI. so like feedback when its time to stop talking. And also stats. Youve spoken 10k words today etc.
Replies
Really love the idea of keeping speech-to-text fully local — privacy-first tools like this are underrated. Does it handle accented English well?
the local mode is the killer feature here — privacy-first dictation without needing an internet connection is huge for people working with sensitive docs. how does the accuracy compare between fast mode and local mode? and is the local model something like whisper under the hood, or a custom model?
B150
@abhisheksinha123 So the accuracy of both the cloud mode and the local mode are nearly identical since they essentially use the same models. The difference is in the amount of AI polish that can be applied. With a cloud model, you can have a lot more degrees of quality versus on device; it's going to be pretty much exactly what you said. All the models we use locally are clearly labeled in your settings section and you can choose between them. And yes, Whisper and Parakeet are the 2 best models we have found.
Local and free is a great combo. Most speech to text tools either need an internet connection or want a subscription for decent quality. How accurate is it compared to something like the built in Mac dictation?
B150
@hiyamojo all of these tools basically use the same models. The difference might be in the instructions in the polish. We use all the top of the line models like Whisper from OpenAI or Parakeet from NVIDIA. Compared to Apple Dictation it 1000x better than that.
a great free solution. I really like it. But it seems that 4000 words are not enough for me. I'll buy the Walkie Pro version
B150
@cherryry1106 happy to offer you 3 month free. Just use the code THANKYOU3 at checkout.
Thank you for offering this kind of local product! I love being able to use my own machine’s processing power to do these things without having to set up an entire development environment just for this.
As a Portuguese speaker, I wanted to ask you if it would be possible to “force” the adoption of a specific word in the dictation, since often, when using a multilingual model, it tries to interpret my Portuguese audio as English, which ends up causing some confusion. Would a small update be feasible that gives the option to “force” the recognition of a specific language?
B150
@llanmelo Yes, if you set your primary language in the app as Portuguese, then it should be able to understand specifically what you're asking. We also do have some routing that happens behind the scenes, so it understands if you're speaking Portuguese or English or something else.
DiffSense
Any recommends for keeping my voice clean? All these talking is making it scrangly 😅 Maybe adjacent business opertunity? :D
B150
@conduit_design what do you mean by clean?
DiffSense
@adam_perlis Voice gets broken when you talk alot. Vibe code side effect 😬
B150
@conduit_design Thanks for sharing but I am still not following your feedback. Do you mind elaborating more. What exactly breaks about the voice when you talk alot. If you can provide a concrete example we can troubleshoot the issue. It could be a number of things including your machine specs, having a certain feature toggled on, running too many programs at the same time, etc.
DiffSense
@adam_perlis No, this is a human thing not a maschine thing. So a human can speak about 10k words before their voice gets brittle, our vocal chords are vibrating at very high freequencies etc. A very active day of coding often involves writing 10k words to an AI. Thats the tension. You can often see it in vibe code videos of devs that talk to their AI. their voice is gone. Esp with people that dont talk alot normally. They have way less than 10k spoken words a day. Maybe no more than 1k words. But there are things you can do. like 10min rest every 50 min. Also maybe analysying brittle ness with AI. so like feedback when its time to stop talking. And also stats. Youve spoken 10k words today etc.