topbanner_forum
  *

avatar image

Welcome, Guest. Please login or register.
Did you miss your activation email?

Login with username, password and session length
  • Sunday December 15, 2024, 4:20 pm
  • Proudly celebrating 15+ years online.
  • Donate now to become a lifetime supporting member of the site and get a non-expiring license key for all of our programs.
  • donate

Author Topic: Tool to facilitate transcribing speech  (Read 5450 times)

CWuestefeld

  • Supporting Member
  • Joined in 2006
  • **
  • Posts: 1,009
    • View Profile
    • Donate to Member
Tool to facilitate transcribing speech
« on: December 21, 2014, 08:37 PM »
My wife is working on transcribing her father's memoirs - he recorded about 39 hours of cassette tapes, which I've digitized for her. She wants to get it all into text. I'm looking for ways to make the job easier.

The obvious solution is something like Dragon software's speech-to-text. But the catch is that the speech is in Mandarin, and so far as I've been able to find, there's no software available to turn Mandarin speech into text.

Short of that, though, can anyone suggest a tool that might help her get through the audio more systematically? Since she's constantly listening to a minute or so, replaying it maybe a couple of times, and then proceeding to the next chunk of text, I think that the beginning of any tool might simplify this looping process.

Another usecase is for when she encounters something she doesn't understand, frequently a name that she doesn't know how to write (Chinese is funny with not being able to write a word if you've never seen its character). She needs to remember these spots and go back over them with her dad, so he can tell her the right characters to use. It might also be useful to store numerous bookmarks for the file, facilitating this process.

Anybody have suggestions I could pass on to her?

4wd

  • Supporting Member
  • Joined in 2006
  • **
  • Posts: 5,644
    • View Profile
    • Donate to Member
Re: Tool to facilitate transcribing speech
« Reply #1 on: December 21, 2014, 10:27 PM »
This might help: Listen N Write

I used it once for transcribing an interview for the Amiga4Ever guys.

You can bookmark by pressing Control+M, (I don't think you can save them though you can take a screenshot of your bookmarks and then jump to time), plus it also has a Smart Pause N Play mode where it pauses when you start typing in the inbuilt editor and when you stop it starts playing after a couple of seconds.

It requires Windows Media Player be installed, (needs wmp.dll), and has its own built in text editor, and is free/donationware.

2014-12-22 15_25_40.png

Renegade

  • Charter Member
  • Joined in 2005
  • ***
  • Posts: 13,291
  • Tell me something you don't know...
    • View Profile
    • Renegade Minds
    • Donate to Member
Re: Tool to facilitate transcribing speech
« Reply #2 on: December 22, 2014, 12:03 AM »
Since you've asked, my own software. (Link is in my sig.)

I've used it many times for transcribing videos, interviews, etc.

The main things that help are:

  • Full and zoomed waveforms to "see" where you are
  • Slowing down the speech
  • Setting loop points
  • Configurable "skip" lengths to quickly navigate
  • Keyboard shortcuts to toggle looping on/off, etc.

On some systems .NET 4.0 Full Profile is required (not the client profile). (This is due to how MS changed where some functions are.)

There's a 15 day trial as well.

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

Freedom is the right to be wrong, not the right to do wrong. - John Diefenbaker

dr_andus

  • Supporting Member
  • Joined in 2012
  • **
  • Posts: 851
    • View Profile
    • Dr Andus's toolbox
    • Donate to Member
Re: Tool to facilitate transcribing speech
« Reply #3 on: December 22, 2014, 04:57 AM »
I use Olympus DSS Player Plus (v. 7), which comes bundled with some of the Olympus digital recorders, in combination with Olympus foot pedal. But I'm not sure if it's possible to buy the software separately. Another option is f4.

The foot pedal frees up your hands for typing, as you can rewind, stop, start with it. Otherwise those functions can be performed by using assigned F keys using the above software.

----

P.S. I just remembered that the Olympus foot pedal might come with its own transcription software called the Transcription Module. But it's a lot more complex and sophisticated than the DSS Player Plus, so I never needed it. But I'm just saying that getting the pedal probably supplies you with prof grade transcription software.
« Last Edit: December 22, 2014, 05:08 AM by dr_andus »

Nod5

  • Supporting Member
  • Joined in 2006
  • **
  • Posts: 1,169
    • View Profile
    • Donate to Member
Re: Tool to facilitate transcribing speech
« Reply #4 on: December 27, 2014, 04:35 AM »
A low budget hardware alternative is a no name usb foot pedal like this in combination with an autohotkey script that sends a pause command to the audio playback device. I've used that when transcribing short audio segments from time to time. For a big transcription project more foot pedals would be useful but for smaller a single pedal may be enough.

Nod5

  • Supporting Member
  • Joined in 2006
  • **
  • Posts: 1,169
    • View Profile
    • Donate to Member
Re: Tool to facilitate transcribing speech
« Reply #5 on: December 27, 2014, 04:43 AM »
The obvious solution is something like Dragon software's speech-to-text. But the catch is that the speech is in Mandarin, and so far as I've been able to find, there's no software available to turn Mandarin speech into text.
Microsoft's real time translation project for Skype will likely support Mandarin in a not so distant future.
For now, the voice function of Skype Translator will only work in English and Spanish, the two most widely spoken languages in the world after Mandarin (it's likely the tech giant will want to launch in Mandarin as well, and position the tool as being for business as well as education). However, in a blog post the company says we can expect "40 plus instant messaging languages" to be up for translations.
http://www.wired.co....kype-translator-tool

While geared toward real time translation of small snippets of speech, I bet there is already work underway by MS and third parties to apply it also to do recorded-speech-to-text-processing. Here is a MS preview video of the Skype real time translation

https://www.youtube..../watch?v=G87pHe6mP0I