Creating a Speech to Text Program with Python

[@danoconnell5292]

I've spent 5 days trying to figure this out knowing nothing literally about nothing and by this I mean virtual assistants you can talk to. With the learning curve I will say I wish I watched your video first. Your the best to learn from so far my man nice job.

[@CSCoach]

That means a lot man. Thanks so much for the feedback.

[@Knot2goodAtIt]

I had nooooo idea this was so straightforward! Thank you! I want to create a translator and I think this is the perfect base!

[@File_corupt]

yo this tutorial is great as I have tried to watch other tutorials as a noob but they talk like I know the stuff but I dont but you explain the stuff to me great

[@goner007]

this worked so well. i never knew it was that easy until you explained it. Hope to learn more from you thanks

[@ymhtpat]

nvm figured it out. for those of you asking about the last step for windows os, go to note pad and type in " C:\Users\[your computer]> " then the first prompt then do the same for the second. save the file as a .txt. it should work from there.

[@BobJoe-lt1is]

What second prompt are you talking about? I wrote that in the notepad and yet my program sometimes outright refuses to work.

[@DevonAIPublicSecurity]

Hey Oscar this was a great tutorial, keep doing the work you are doing and this was very clear and it made perfect sense.

[@CSCoach]

Thanks for the feedback. It means a lot. Going to ramp up the channel with more content in the coming months :)

[@artetridimensionale]

ok so it's a bit simple, the problem is if you stop talking and you have to make sure the text is justified and doesn't end up at the end of a line every time you interrupt yourself

[@ColinTimmins]

Cool stuff! I have struggled so much with language as I’m extremely dyslexic. ChatGPT has opened up the door 🚪 for me! 😊

[@CSCoach]

That's awesome :D So glad ChatGPT is able to help you :)

[@ymhtpat]

On the last step I'm hung up. I'm on the latest windows os, Are you running the touch output.txt and tail output on macs equivalent of command prompt? I ran it on mine and came up with " 'touch' is not recognized as an internal or external command, operable program or batch file." where did I go wrong?

[@joyaljijimon3419]

same here bruh

[@poisoned_durian8]

im using windows im not so sure if the problem is the commands "echo. > output.txt and type output.txt | more"

[@pattuchiitu8978]

Thank you for the tutorial. its works now 👋

[@MrScgaming28]

Can you tell me how to do the last step of making that output file and tailing the outputs in a windows OS?

[@anderson3889]

did you find a way

[@arvindh13]

In windows it will record your audio & print it to the notepad, but you need to restart your notepad to see it.

[@sandrasajeev8640]

@@arvindh13 Hey! I'm also stuck at this step. Can you pls explain like where do I've to put the commands to touch and tail the output?

[@mmjuuno]

how do i see the appended msgs in the terminal? it worked and I could open the output txt file and see what i was saying, but I'm not sure how to see it realtime ig (using Windows 10 and the windows subsystem for linux to install kali linux, and then opened bash in cmd). also when i stop the script (using pycharm) it spits back a bunch of errors and I'm not sure why, or at least i think they're errors

[@CSCoach]

I did this by running the tail command in another terminal. Though, you could also change line 43 to be print(text) rather than print("Wrote text")

[@kavito5947]

@@CSCoachwhats the name of the terminal you used?

[@mohitpandya_2228]

This is just working fine for the first run. After the first run the text generated takes about 10 mins to get to the output. How can i fix that thing and make it just as fast as the first run

[@Illogical.]

I need a thing that does a few steps less than what this does. I only want it to record phonetics, maybe spacing between words, maybe intonation, and maybe recognize when I'm pausing to remember a word.

[@loisisnel2955]

Thanks for the course, but does anyone know how can i do it by using an audio file and not a voice recognition ?

[@anirudhsrisai3397]

Think of integrating it with chatgpt and getting answers for it just by using the text which we have received from speech to text

[@CSCoach]

Haha, did you see this video I made :) I believe I do what you suggested in the comment. kzbin.info/www/bejne/eHbaZnh8pqh2e6s

[@tugpsx640]

This is great, thanks for sharing you tips and tricks.

[@CSCoach]

You bet!

[@KNOCKOUT-t7e]

@@CSCoach what is the name of the app u used in the video ??

[@Yajnco]

My language Hmong is not available for voice to text and I have been searching for programs or ways to show me how it's done, but can't find any. Not sure if Python is the solution. Frustrating of not knowing where to start.

[@alexanderkartvelian4274]

You are awesome! It would be great if you teach us how to send generated text from recognizer into "Text to speech" . thanks for the video!

[@SirMrMystery]

he already did something like that in which he created a jarvis like program kzbin.info/www/bejne/eHbaZnh8pqh2e6ssi=wEeSWa18kFPppBUF

[@aotrakstar]

Hi coach. do you think maybe using openai's whisper will have a more accurate outcome for transcribing speech?

[@CSCoach]

I found the python library to be enough personally. Though, I would imagine whisper to be better given that it's made by openAI. I'd expect it to be a better trained AI

[@hjoseph777]

You can use whisper offline. I suggest faster-whisper or whisper.cpp

[@abaizkhan4963]

Any git repo link?

[@snaxsammy6472]

Hi Coach! MrScgaming28 and I, from the comment section seem to have the same issue, could you re-explain how to do the last step of making that output file and tailing the outputs in a windows OS?

[@arvindh13]

Create a text document & copy the source --> paste it in the program.

[@story-4-you-man]

@@arvindh13 can you explain this? program?

[@JamesDonkor-edu]

hey great video! would this work for other languages?

[@danielcasas7343]

I also hace the same question

[@workstation-s2n]

Could you please post a link to the full file

[@jimjones26]

I am going to work on adding a "trigger" word that will let the program know I want to act.

[@CSCoach]

That's cool!! If you get that to work, I'd love to know how

[@mosawarjamshady2989]

Hey did you end up figuring this out?

[@fransuacordero5407]

one question, this works only for english lenguage? or can be used with other languages?

[@Quagik]

its saying AttributeError: module 'speech_recognition' has no attribute 'Recognizer'

[@ernstb1234]

It keep telling me Zach command not found when I try installing the dependency please help

[@mohamed_Sultan1504]

Does anyone know how to out it in a user-friendly app so anyone can use it easily?

[@Ddetektiv]

how do you get terminal? When I clicked on terminal, new shows me and there was PS C:\Users\[my name]>, but as I can see, in ur it's not. Any way how to fix it?

[@snaxsammy6472]

At the top bar of the terminal where it says "problems" click on the drop down arrow next to the "+" in the further right of that bar, then click on "command prompt", you should be able to run the code through there. At least thats what I did and it worked for me.

[@Kenoki-yi3gf]

i am doing this on windows os and i cant seem to find where the text is saved.... can someone please help

[@Hbdisaster_28]

For me whenever i say something, the terminal shows wrote text but idk where it writes the text(im on windows btw) can anyone plz help

[@arpitv2003]

it would be creating a text file with name "output.txt" and writing into it

[@Hbdisaster_28]

Thank you

[@sicfrydred]

what program did you use to do this?

[@CSCoach]

Did it in python :)

[@Bartdu59Gaming]

He used "VS Code" and the language used for the program is Python.

[@kavito5947]

@@Bartdu59Gamingwhats the name of the terminal he used?

[@usus8420]

hi it's only desktop solution ?

[@hjoseph777]

You did Not show much also where is the Git repository link. Smartie

[@Lunaraa0]

Hi oscar does it works for french

[@OmarAbdelrazek-tn9du]

I still don’t understand how to download the libraries, I have a Mac m2 and documentation u gave didn’t help me. So can u or anyone else help me?

[@felixforster5836]

You have to type in the terminal "pip3 install ...."

[@adrio1569]

How can we add languages to it?

[@arvindh13]

You need pre designed language libraries or you need to make a new one with a lot of data.

[@mohamed_Sultan1504]

A real king

[@user-db4nm2rp2w]

if i want to do only speech to text, then I don't need to install pyaudio, right??

[@arvindh13]

Yes, you need Pyaudio & Speech recognition fork library

[@i.a_n_i_l_k_u_m_a_r]

How to stop recording the voice?

[@suissegarantiegaming2100]

Could this also work with a other Language?

[@CSCoach]

Programming language? Or the text is outputted to a different language? Either way I'm pretty sure the answer would be yes :)

[@KNOCKOUT-t7e]

what app he used in this video??

[@stylloofdreams975]

Visual Studio Code

[@KutuluTuk]

i couldnt import pyttsx3 it says unused and alot of error there 😂

[@arvindh13]

you need to install it, it's not an inbuilt library.

[@falcongold2024]

why import pyttsx3? it is never used?

[@arvindh13]

It's not required yes.

[@manchishirisha8013]

where are u writing the code

[@ruzix_yeager4938]

vscode

[@WolFX_FPS]

I have an error with the speech_recognition library: r.adjust_for_ambient_noise(source2, duration=0.2) TypeError: Recognizer.adjust_for_ambient_noise() missing 1 required positional argument: 'source' Any ideas?

[@arvindh13]

You need to provie the source before that, please check the program again and if it does not work, install Speech recognition fork library.

[@gleful262]

got this error, fixed it by adding() to Recognizer

[@thunderchills640]

thanks

[@dungnguyen-si2sz]

i am getting trouble with this error: 1 error generated. error: command '/usr/bin/clang' failed with exit code 1 [end of output] note: This error originates from a subprocess, and is likely not a problem with pip. ERROR: Failed building wheel for pyaudio Failed to build pyaudio ERROR: Could not build wheels for pyaudio, which is required to install pyproject.toml-based projects I cant install the two last kind of stuffs you gave us, hope you reply soon, thank you for that nice work sir.

[@mikethompson6455]

Getting same error. Hope someone can help

[@mikethompson6455]

Did this. I have Mac OS so If you're using macOS, you can install the necessary tools using Homebrew. First, make sure you have Homebrew installed, then run: brew install portaudio After installing portaudio, you can try installing PyAudio again using pip3 install pyaudio.

[@gleful262]

had this issue on widows fixed it by running as admin

[@odogbolahan8148]

my name is oscar too haha

[@hebataha5197]

Does this work offline ?

[@brianmckeown70]

^

[@hjoseph777]

Your screen not very clear

[@soham0726]

source code

[@phpsolutioncode9309]

I continue your work ! thank you! for your video was essential for my Idea. kzbin.info/www/bejne/jWetlHxvhtxrl7c