Welcome Guest.   Make a donation to an author on the site December 20, 2014, 12:42:09 AM  *

Please login or register.
Or did you miss your validation email?


Login with username and password (forgot your password?)
Why not become a lifetime supporting member of the site with a one-time donation of any amount? Your donation entitles you to a ton of additional benefits, including access to exclusive discounts and downloads, the ability to enter monthly free software drawings, and a single non-expiring license key for all of our programs.


You must sign up here before you can post and access some areas of the site. Registration is totally free and confidential.
 
Learn about the DonationCoder.com microdonation system (DonationCredits).
   
   Forum Home   Thread Marks Chat! Downloads Search Login Register  
Pages: [1]   Go Down
  Reply  |  New Topic  |  Print  
Author Topic: mp3 audio to text  (Read 4032 times)
Contro
Participant
*
Posts: 2,186

View Profile Give some DonationCredits to this forum member
« on: June 11, 2011, 06:32:11 AM »

mp3 audio to text

I supposed this is very difficult.
I have a mp3 audio previously recorded. Now I want to convert to text.

Additionally may consist in the discussion between two voices.

What tools can I use to make this work easier ?

Best Regards

Logged
Renegade
Charter Member
***
Posts: 11,964



Tell me something you don't know...

see users location on a map View Profile WWW Give some DonationCredits to this forum member
« Reply #1 on: June 11, 2011, 06:35:00 AM »

There are quite a few tools out there to help. Search for "speech to text". Dragon Naturally Speaking is one. There are more. Most are not free, though there are some free ones. I've only looked into it from a development perspective, so I can't really recommend much on the consumer side. That should point you in the right direction though.
Logged

Slow Down Music - Where I commit thought crimes...

Freedom is the right to be wrong, not the right to do wrong. - John Diefenbaker
Contro
Participant
*
Posts: 2,186

View Profile Give some DonationCredits to this forum member
« Reply #2 on: June 12, 2011, 01:18:47 PM »

Are you sure Dragon Naturally Speaking is one for transcripting an audio to text.?
DNS is used to dictate from voice to text. But in real mode until I know. Recognizing the voice you have trained before.
Logged
Renegade
Charter Member
***
Posts: 11,964



Tell me something you don't know...

see users location on a map View Profile WWW Give some DonationCredits to this forum member
« Reply #3 on: June 12, 2011, 02:15:12 PM »

Are you sure Dragon Naturally Speaking is one for transcripting an audio to text.?
DNS is used to dictate from voice to text. But in real mode until I know. Recognizing the voice you have trained before.


They have STT solutions. Not sure if they have consumer level ones there though... But it exists. Samsung bada will have it soon in v2.0 or so.
Logged

Slow Down Music - Where I commit thought crimes...

Freedom is the right to be wrong, not the right to do wrong. - John Diefenbaker
Contro
Participant
*
Posts: 2,186

View Profile Give some DonationCredits to this forum member
« Reply #4 on: June 12, 2011, 07:08:13 PM »

Do you really know one to convert an audio to text file ?
Logged
Contro
Participant
*
Posts: 2,186

View Profile Give some DonationCredits to this forum member
« Reply #5 on: June 13, 2011, 06:17:30 PM »

If I found one I will tell. I'm trying with no good results.
May be a type of experimentating software.
Logged
Curt
Supporting Member
**
Posts: 6,363

see users location on a map View Profile Give some DonationCredits to this forum member
« Reply #6 on: June 13, 2011, 07:22:29 PM »

http://shop.nuance.com/st...B/pd/productID.202232300/

Quote
The most accurate transcriptions were of recordings I made in my office, windows closed, with background noise at a minimum. The transcriptions of those recordings were 85 to 95 percent accurate.

5 years ago there were still several STT applications. Most seems to have given up by now, maybe because they can't compete with the Dragon / MacSpeech Dictate or Adobe Audition CSS 5.5. Google Translation includes STT, but it is not yet convincing. Also, Chrome makes Text-to-speak & Speak-to-text become a part of HTML-5, but it is still primitive. If you already have a iPhone 5, it may be the most affordable way to go!

Quote
Nuance speech-to-text references baked into iOS 5 internal settings | 9 to 5 Mac

<a href="http://www.youtube.com/watch?v=3eD_HDWSDPM" target="_blank">http://www.youtube.com/watch?v=3eD_HDWSDPM</a>
Logged
Contro
Participant
*
Posts: 2,186

View Profile Give some DonationCredits to this forum member
« Reply #7 on: June 13, 2011, 07:24:55 PM »

Uffffff Curt.
It's not the same.

The point is transcript an audio with one or more voices to text

audio file to text file

NO SPEECH TO TEXT

Best Regards
Logged
Contro
Participant
*
Posts: 2,186

View Profile Give some DonationCredits to this forum member
« Reply #8 on: June 17, 2011, 09:43:35 AM »

I close this post with not results.
It's a software not developed until this moment.

Best Regards
Logged
tanvi12
Participant
*
Posts: 3

View Profile Give some DonationCredits to this forum member
« Reply #9 on: August 08, 2013, 05:23:25 AM »

Hello Contro ! Software are there like Dragon Natural Speaking and more. But you can also opt for online speech to text services like www.speechtotextservice.com. These services really work great.
Logged
patteo
Charter Member
***
Posts: 436


View Profile Read user's biography. Give some DonationCredits to this forum member
« Reply #10 on: August 08, 2013, 10:13:12 AM »

I use Dragon NaturallySpeaking 12 Premium.

Yes, Dragon NaturallySpeaking allows you to Transcribe a recorded audio file. See the attached image.

However, let me warn you that the results are not what you would expect or hope for. There is likely to be several errors, missed words, mis-recognized words and if you want a reasonably good transcription, I think you would be disappointed.

I have tried it several times to get a reasonably good MP3 to be processed into text by Dragon NaturallySpeaking without human intervention, but the results contain far too many errors and the subsequent correction process to tidy it up simply takes too long.

The reason for this is multifaceted; the voice model of the speaker may not be available to you (ie - it is not trained to his voice), the MP3 quality may have a lot of background noise, or the microphone was not of a good enough quality or it may be too far away from the speaker's mouth, or he may be speaking too close or too directly into it and it's capturing his breathing sounds and he may slur his speech, or not pronounce words carefully  (ie the speaker is not speaking specifically for speech transcription). There are many factors that contribute to accuracy or lack of accuracy.

Although Dragon NaturallySpeaking has really come of age and doesn't really require that much training in version 12, it does require that you enunciate the words reasonably carefully, without speaking too fast but rather speaking naturally and of course you need a fast laptop as it is quite CPU intensive.

Those people who have used speech transcription will tell you that the most accurate and efficient and effective method is really listen and read back. What that means, simply is that you listen to what the speaker is saying on the MP3 for example and then you speak into a good quality microphone, ie you act as a go between. If you do it this way, then Dragon NaturallySpeaking will give you outstanding results.

However, please take note, it takes approximately about an hour to transcribe 10 minutes of MP3 if you used the listen and read back method.
It is also Critical to make sure that you use a high quality microphone that is attached to a good quality sound card or a good quality USB sound pod.

Another way, is to pay for someone to transcribe for you. Just Google for transcription services.

After all is said and done, you perhaps may still want to just try using Dragon NaturallySpeaking to process the MP3 to see the  whether the results are adequate enough for you.


* TranscribeFromMP3.JPG (39.62 KB, 621x400 - viewed 71 times.)
Logged
Renegade
Charter Member
***
Posts: 11,964



Tell me something you don't know...

see users location on a map View Profile WWW Give some DonationCredits to this forum member
« Reply #11 on: August 08, 2013, 10:56:50 AM »

I'm not sure if Google is still doing their voice to text on YouTube anymore - I've not seen one in a while.

Andrew and I did an interview with Adam Smith on the problem. He's working on a solution, but it has a radically different approach than DNS. However, Andrew took down the site where the interview was, so it's no longer available there, but it looks like it is on iTunes:

https://itunes.apple.com/...eport-podcast/id456825773

Adam explains the problem with the current approaches quite well.
Logged

Slow Down Music - Where I commit thought crimes...

Freedom is the right to be wrong, not the right to do wrong. - John Diefenbaker
Contro
Participant
*
Posts: 2,186

View Profile Give some DonationCredits to this forum member
« Reply #12 on: August 08, 2013, 06:11:03 PM »

 Kiss
I am going to learn.
Best Regards
Logged
IainB
Supporting Member
**
Posts: 4,919


Slartibartfarst

see users location on a map View Profile Give some DonationCredits to this forum member
« Reply #13 on: August 09, 2013, 11:36:19 AM »

Do you actually need a transcript of speech/discussion, or simply the ability to search for a phrase and the point in time on the recording when that phrase occurs?
If you just want the latter, then slinging the mp3 or wav file into MS OneNote will get it indexed and searchable.
As discussed in this thread: Microsoft OneNote 2007 - some experiential Tips & Tricks
Logged
Contro
Participant
*
Posts: 2,186

View Profile Give some DonationCredits to this forum member
« Reply #14 on: August 10, 2013, 04:45:01 AM »

It's is the transcript of speech/discussion.

 tongue
Logged
CleverCat
Supporting Member
**
Posts: 1,129


Cat's Are Fun!

see users location on a map View Profile Give some DonationCredits to this forum member
« Reply #15 on: August 11, 2013, 01:50:06 AM »

I use Ivona products -excellent!

link
Logged

If you need help - JUST ASK!
Contro
Participant
*
Posts: 2,186

View Profile Give some DonationCredits to this forum member
« Reply #16 on: August 11, 2013, 07:59:57 AM »

I use Ivona products -excellent!

link

 embarassed

I am meaning audio to text, not text to speech......
Logged
IainB
Supporting Member
**
Posts: 4,919


Slartibartfarst

see users location on a map View Profile Give some DonationCredits to this forum member
« Reply #17 on: August 11, 2013, 08:22:25 AM »

You might find something useful here: http://duckduckgo.com/?q=Wave+to+Text+converter
Logged
Contro
Participant
*
Posts: 2,186

View Profile Give some DonationCredits to this forum member
« Reply #18 on: August 11, 2013, 08:45:11 AM »

Running to learn
Logged
CleverCat
Supporting Member
**
Posts: 1,129


Cat's Are Fun!

see users location on a map View Profile Give some DonationCredits to this forum member
« Reply #19 on: August 13, 2013, 04:36:03 AM »

I use Ivona products -excellent!

link

 embarassed

I am meaning audio to text, not text to speech......


Oops - apologies!  embarassed
Logged

If you need help - JUST ASK!
Pages: [1]   Go Up
  Reply  |  New Topic  |  Print  
 
Jump to:  
   Forum Home   Thread Marks Chat! Downloads Search Login Register  

DonationCoder.com | About Us
DonationCoder.com Forum | Powered by SMF
[ Page time: 0.037s | Server load: 0.08 ]