My initial impressions of AtTC (Audio to Text Converter) after some testing:

1) I wanted to dictate to the computer & have it transcribe it, so in my first attempt I tried the "Transcription" mode. At first I was getting nothing. I noticed that in that mode it defaults to looking for audio from the system speakers rather than the microphone. EVERY time. So to use that mode you have to start it, then click on the microphone, every time. Why is it looking for audio from the system speakers?

Then I tried the "Record" function. As far as I can tell both modes work exactly the same except Record goes to the mic by default.

My take is that the "Transcription" mode is for transcribing things recorded previously. This may be useful for students who have recorded lectures; people who have recorded meetings; etc. But as AtTC exists only to transcribe, it would be clearer to label the functions as "transcribe from mic.", " transcribe recording", etc.

2) When I told AtTC to use the mic instead of the system speakers it still didn't work. Turned out that I had to go several levels into audio settings to find out my mic had been muted. Then it started hearing me & transcribing, So if it doesn't work for you, definitely check to see that the problem isn't somewhere in your system's audio settings.

3) Both Record and Transcribe modes appear to make audio recordings of what you're saying & when you go to export the file, both give you the option of keeping or deleting the audio. It's not bad to be able to optionally keep recordings of dictation, but the program defaults to keeping everything (audio, notes, etc.) If you just want to dictate to the mic & have the computer transcribe it to a text file, by default the program wants to save way more data than you need.

4) AtTC defaults to what it wants to default to & there does not appear to be any way to change those default options. So every time I want to export a file, I've got to change it from .pdf (default) to .txt [other optional export formats are .html and .docx].

Next you have to un-click the (by default turned on) export options of:
- Show speaker
- Show timestamps
- audio file
and
- highlights & comments

Again, if you just want to dictate & have your voice turned into text which you would keep, this is all needlessly complex.

5) The actual audio to text part.

The program does a fair job at this. I'd say it's about 80-85% accurate. But that means you have to proofread everything carefully & fix about 15-20% of what it has converted. At that point, you've spent so much time between the dictation and editing that you may be better off just typing it in the first place.

AtTC is 100% intolerant of any brief pauses in your dictation. Not even taking a breath is allowed. You must dictate in an unbroken stream because as soon as you breathe it will start a new line of text.

If you are dictating with longer pauses where you may stop for a second or more to think about what you're going to say next AtTC goes beyond starting a new line. It determines you are a whole new speaker and reflects that in the transcription:
Speaker 1: Today I went for a walk.
Speaker 2: The sun was shining and a gentle breeze moved the leaves in the trees.
Speaker 3: I stopped to watch a group of kids throwing and chasing a frisbee.
Speaker 4: Two squirrels took turns chasing each other around a tree.

There is no way to turn off this multiple speaker function, or change the duration of allowable pauses before a new line or new speaker is automatically selected. When you export the final transcribed project you can unclick the options for show speaker & show timestamps, but there will still be line breaks for every new sentence.

Conclusion: For those who are slow hunt & peck typists, AtTC could be useful, but for touch typists with any degree of competence I don't see a significant advantage in using this to dictate to your computer over using the keyboard. I suspect this is where Dragon Naturally Speaking was a decade ago, but I never did much more than play around with that program & don't know how it compares today.

Being able to change the default settings would be a big improvement as well as being able to disable the program automatically starting a new line / new speaker after every pause.

If I need to transcribe voice-to-text, at this point I think I'll do better using my (Android) phone as a dictation machine, sending dictated audio to myself through text messages or Fb Messenger.

That said, I appreciate EaseText providing an opportunity to evaluate your product and as always, thanks to the GotD team for bringing this offer to us.

Installed easy.
Tested on mp3 - it seemed to work to some extent - translate speed is kind of a fast talking speed.
But when I stopped and exported the result to text, was null text file?

As soon as I downloaded, installed, and launched the software, I got the following pop-up:

"This version of EaseText software has become obsolete. Please download the latest version to install on your computer."

The only way to close it was to click OK, then it closed the program, opened my browser & went to a page to download "easetext-audio-to-text-converter.exe"

The GotD downloaded setup file is 77,859 KB (unzipped) and dated 1/4/23

The new setup file is 77,860 KB and dated to the moment of download.

When I tried relaunching the program it did the same thing. I could not register as clicking the OK on the pop-up window closes the program & goes to the auto-download page. So I re-installed with the new 3.1.3 version. Entered the provided user name and License key & it registered successfully.

Note that license expires at the end of the calendar year: 12/31/23, so it's a 10½ month license rather than a year, but that's still a $31 value, which is appreciated. I look forward to seeing how well the voice recognition compares to that on my Android phone.

Thank you GotD and EaseText.

does this work live ?
Meaning we can connect a microphone, and it types what you say ?
Also can you train it by reading out text that it presents for you ?
I don't think Google Docs has training, and I wish it did.

I'm trying to install it in a second PC. When I did, it needs another mail for the registration key which I provided but it says that there are too many request from my IP address, so, I cannot get a second key to register the software :'-(

Installed easily, but I have to say the text created when I spoke into the mike was nothing short of hilarious!

Microphone settings can't find it, and i have 8 input on my pc

Not impressed by the accuracy. I tested a clean audio mp3 clip of JFK's New Frontier speech: "A new frontier is here whether we seek it or not. Beyond that frontier are uncharted areas of science and space, unsolved problems of peace and war ..." The EaseText transcript was: "a new frontier is a where whether we seek it or not beyond that pronto yeah or uncharted area of science and face on fall problem the these more on call good province..." Maybe it will work better for you.

Unfortunately useless. I checked it with a non-English audio, and the translation looks like people had hiccups. Only a small part of the conversation was detected and converted.

Hi, quick double-check: Is this supposed to be a *full* one year license?

I just registered the tool. It is for the Personal Edition, but the expiration date is 2023-12-31 and not 2024-02-13 , I expected.

Not upset, just looking for clarity. Thank you!

Is this one better than what Google has in their Google Docs?

Limit languages, is there a possibility to add more languages?

Oh, the license key came. It was just late.
Thanks, EaseText and GOTD!

What happens after the one year of this Giveaway? Does the program stop working?

I tried to install any language different from English, but this requires some source "model" file.
Where are they available?