Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Transcribe.cpp

workshop.cjpais.com

Post reply on HN
Print view View original

51–60 of 180 posts

Re: Transcribe.cpp

#51

Post by mmmmbbbhb »

abdullahkhalids wrote: post #46

For anyone looking to build on top of this. I have tried a few different STT systems, and they accurately capture what I am saying. Unfortunately, they don't support the reasonable workflow I want to open an office document, for example, and start talking. And I want the software to continuously type what I am saying at the cursor with minimal latency. The continuous part is crucial. Many software will paste whatever…

You know English doesn't work like that. The word you're saying only becomes clear with the surrounding context. Eg, 'there' vs 'their'.

Re: Transcribe.cpp

#52

Post by primaprashant »

abdullahkhalids wrote: post #46

For anyone looking to build on top of this. I have tried a few different STT systems, and they accurately capture what I am saying. Unfortunately, they don't support the reasonable workflow I want to open an office document, for example, and start talking. And I want the software to continuously type what I am saying at the cursor with minimal latency. The continuous part is crucial. Many software will paste whatever…

Totally understandable, but I’ve found that software that transcribes everything after I finish recording actually works better for me. I’ve tried both kinds, and systems that continuously type what I’m saying distract me from completing my thought. I end up reading what’s being typed and noticing transcription mistakes instead of focusing on what I’m trying to say.

I often prefer to dictate everything in my head about a particular thing for 5–10 minutes and then go through it afterward. I find that much more useful because it doesn’t break my thought process the way continuous transcription does.

Re: Transcribe.cpp

#53

Post by electronstudio »

abdullahkhalids wrote: post #46

For anyone looking to build on top of this. I have tried a few different STT systems, and they accurately capture what I am saying. Unfortunately, they don't support the reasonable workflow I want to open an office document, for example, and start talking. And I want the software to continuously type what I am saying at the cursor with minimal latency. The continuous part is crucial. Many software will paste whatever…

This is what I attempted with https://github.com/electronstudio/low_latency_dictation

However the accuracy of the real time models is poor, so I did a second pass with a higher accuracy model before committing the text.

Re: Transcribe.cpp

#54

Post by nilslindemann »

mmmmbbbhb wrote: post #51
abdullahkhalids wrote: post #46

For anyone looking to build on top of this. I have tried a few different STT systems, and they accurately capture what I am saying. Unfortunately, they don't support the reasonable workflow I want to open an office document, for example, and start talking. And I want the software to continuously type what I am saying at the cursor with minimal latency. The continuous part is crucial. Many software will paste whatever…

You know English doesn't work like that. The word you're saying only becomes clear with the surrounding context. Eg, 'there' vs 'their'.

It may be interesting to have it immediately insert the words, even if they are wrong, and when a sentence is finished, replace what has been written with the final corrected sentence.

Re: Transcribe.cpp

#55

Post by catmanjan »

abdullahkhalids wrote: post #46

For anyone looking to build on top of this. I have tried a few different STT systems, and they accurately capture what I am saying. Unfortunately, they don't support the reasonable workflow I want to open an office document, for example, and start talking. And I want the software to continuously type what I am saying at the cursor with minimal latency. The continuous part is crucial. Many software will paste whatever…

You used to be able to do this with dragon naturally speaking (don’t remember if that was it’s exact name) 10 ish years ago

Re: Transcribe.cpp

#57

Post by dostick »

mmmmbbbhb wrote: post #51
abdullahkhalids wrote: post #46

For anyone looking to build on top of this. I have tried a few different STT systems, and they accurately capture what I am saying. Unfortunately, they don't support the reasonable workflow I want to open an office document, for example, and start talking. And I want the software to continuously type what I am saying at the cursor with minimal latency. The continuous part is crucial. Many software will paste whatever…

You know English doesn't work like that. The word you're saying only becomes clear with the surrounding context. Eg, 'there' vs 'their'.

Model should be able to understand where logical sentence ends, to stop buffering, and optionally rewrite some of the test that has already been output.

Re: Transcribe.cpp

#58

Post by samplifier »

dostick wrote: post #57
mmmmbbbhb wrote: post #51

Earlier quoted context omitted.

You know English doesn't work like that. The word you're saying only becomes clear with the surrounding context. Eg, 'there' vs 'their'.

Model should be able to understand where logical sentence ends, to stop buffering, and optionally rewrite some of the test that has already been output.

IIRC that is exactly how Dragon Naturally speaking did it decades ago.

Re: Transcribe.cpp

#59

Post by atonse »

mmmmbbbhb wrote: post #51
abdullahkhalids wrote: post #46

For anyone looking to build on top of this. I have tried a few different STT systems, and they accurately capture what I am saying. Unfortunately, they don't support the reasonable workflow I want to open an office document, for example, and start talking. And I want the software to continuously type what I am saying at the cursor with minimal latency. The continuous part is crucial. Many software will paste whatever…

You know English doesn't work like that. The word you're saying only becomes clear with the surrounding context. Eg, 'there' vs 'their'.

But this is still possible to do if you track the whole run of text. You could replace all of it each time so it LOOKS like it’s streaming but earlier words also change. I’m hoping the streaming models do this eventually.

I believe the built-in iOS dictation already does this.

Re: Transcribe.cpp

#60

Post by iezepov »

eventualcomp wrote: post #16
boplicity wrote: post #5

Earlier quoted context omitted.

>This makes these projects so much more trustworthy and easier to approach: >> Were any of the words here written using AI? Nope. They came from my mouth or my fingers. I have to push back on this a bit, as I believe (quite strongly) that we're shaped by the tools we use; text-to-speech LLMs are still LLMs, and generally their mistakes are shaped by the expectations inherent in their training. This, in turn, shapes t…

Isn't this like saying "my words are not really my own when I speak to my family, because I know my father is a non-native English speaker and hard of hearing so I try to use words which are well enunciated and are few in syllable count"?

You can take it one step further! As Tyutchev wrote, "A thought once uttered is a lie." [1] Speech is a projection of a thought, and a lossy one. So no matter who is the listener, the speaking/writing does affect the thinking. Though comment on LLM transcribing is spot on.

1. https://www.poetryloverspage.com/poets/tyutchev/silentium/li...

Post reply on HN