Jump to content

adding timespans to imported audio transcript


Recommended Posts

Hello,

 

I'm new to NVivo on a 30 day trial, and I'd like to use it to work with audio transcripts of group meetings. I have a very rough partial transcript that I'd like to use as the starting point for a complete transcript. I've imported my rough transcript to the audio file, but it doesn't have any timespans. I need a timespan for each of the speaking turns in the group conversation, and I'm finding it terribly clumsy to delineate each interval on the audio file and assign it to a row -- the selection keeps disappearing on me, the audio cursor hops around, it's hard to start on the next turn directly where I left off, and I feel like it takes 5-10 minutes of fiddling just to add a timespan to one turn (out of over 100). And once a turn has a timespan, it moves to the bottom, so I'm constantly having to scroll up and down the whole document to get a direct correspondence between turns. It seems terribly inefficient. I can't find any info on this in the help files. Any ideas on how to do this more efficiently?

 

Thanks!

 

Ann

Link to post
Share on other sites

No one has responded so I guess there's no easy fix here. This is much easier in the audio/video-based program, Transana, which works with simple timestamps. But then I can't link to my other source documents. :( Please let me know if anyone has a solution.

Link to post
Share on other sites

Hi Ann,

 

Sorry you are having trouble with your audio transcript.

 

Maybe it would be more efficient for you to add the timespans in the transcript before importing it.

 

If your original transcript is in table form, you could add the timespans into the first column - check the attachment to see what the document should look like before you import it.

 

You can find out more about importing transcripts in the help topic Import audio or video transcripts.

 

Hope this helps!

 

Cheers,

Kath

timespan.jpg

Link to post
Share on other sites

Thanks for your reply, Kath!

 

What I figured out today is that the easiest thing is to complete the transcription in Transana, which makes it a whole lot easier to add timestamps, and then import the transcript produced by Transana (with timestamps) into NVivo. It takes some minor reformatting of the timestamps (taking out the parenthesis) but it converts the timestamps into timespans just fine. I wish there was a way to do this all in one program, but I find transcription to be a whole lot easier in Transana, because it's just one continuous document. The timespan windows in NVivo make transcription more clumsy, in my opinion. But Transana clearly can't do the kind of hierarchical coding and data analysis that NVivo can do, and it can't link to other kinds of sources (text, photos), so I'll have to work with them both. I hope that NVivo improves its transcription tool in the future, although I do realize this is a big advance over previous versions of the program.

 

Ann

Link to post
Share on other sites

Ann,

 

I utterly agree with you that the transcription feature of NVivo is a good concept but with a very clumsy, poor, and inefficient implementation. IT is also quite difficult, you should know, to code to transcripts. In some cases codes applied to multiple lines of transcript count more than once (i.e., once for each row), and coding stripes have to span an entire row, no more, no less. IN sum, this is a very bad implementation of the transcription process, especially compared to what competitors offer (like MaxQDA). However, I would suggest one method that Nvivo really has developed far better than the competition (at least at this point), and that is the ability to code directly to the audio files. I have done this to economize my workflow and go more directly from data collection to analysis. You can apply codes to the audio, mark sections for transcription, and even develop a code book based directly on the audio stream/timeline. I use a mix of this and analyzing transcripts, although I gave up on Nvivo's transcript features. I just import my transcripts as document and then I create links between the relevant sections of audio and the transcription. This is a temporary work-around, but hopefully QSR overhauls its implementation of transcript-audio interface in future versions.

 

-Jacob

 

 

Thanks for your reply, Kath!

 

What I figured out today is that the easiest thing is to complete the transcription in Transana, which makes it a whole lot easier to add timestamps, and then import the transcript produced by Transana (with timestamps) into NVivo. It takes some minor reformatting of the timestamps (taking out the parenthesis) but it converts the timestamps into timespans just fine. I wish there was a way to do this all in one program, but I find transcription to be a whole lot easier in Transana, because it's just one continuous document. The timespan windows in NVivo make transcription more clumsy, in my opinion. But Transana clearly can't do the kind of hierarchical coding and data analysis that NVivo can do, and it can't link to other kinds of sources (text, photos), so I'll have to work with them both. I hope that NVivo improves its transcription tool in the future, although I do realize this is a big advance over previous versions of the program.

 

Ann

Link to post
Share on other sites

Hi Jacob,

 

Thanks for your feedback - improving the implementation of audio/video transcription is definitely on our radar. Just to be sure, I've raised your specific concerns with the development team. Good to hear that coding directly on the audio is working for you - I like your work-around too.

 

Cheers,

Kath

Link to post
Share on other sites

Thanks, Jacob! It's helpful to know that this wasn't just my fault on the clumsiness of the transcription tool. I've given up on trying to transcribe directly and farmed out the transcripts to a professional transcriber instead. She'll include timestamps, and then I'll just upload it to the audio. I'll look out for that problem in terms of coding to the uploaded transcript. So you're just working with the transcript as a separate text and linking specific audio segments to that?

 

Ann

Link to post
Share on other sites

Also Kath -- for future versions, I'd suggest dropping the text boxes on the transcript, which seem to jump around the document out of your control, given odd numbering and ordering features. Just have a continuous transcription document with timestamps rather than timespans. That's MUCH easier to work with, in my experience. I lost several workdays trying to work with this.

Link to post
Share on other sites

This post may also be of interest as it describes a method for creating transcripts using Transcriber 1.5.1 and then importing the transcript along with timestamps and audio files into NVivo.

 

Transcriber has recently been replaced by a completely new and much improved transcription program, TranscriberAG. I'm not sure how one would go about exporting transcripts with timestamps for import into NVivo using the new version. Another time-aligned transcription program worth looking at is XTrans. These tools, along with Transana, are much better at transcription than any of the available QDA packages that support audio/video data with time-aligned transcripts. I think it would be to the advantage of programs like NVivo to provide import directly from the transcription formats supported by these programs.

 

Alan.

Link to post
Share on other sites

Thanks, Alan! I found that using Transana, if I make the timestamps visible and then export the transcript in rich text format, and then take out the parenthesis around the timestamps, the transcript and audio will import into NVivo with the timespans between timestamps intact. A bit time-consuming as a work around.

 

I agree with you about the urging NVivo to support import from transcription formats in Transcriber, Transana, etc. It would be especially useful if the audio codes in Transana could be kept intact upon import to NVivo, so that the transcripts and audio could be linked to the more sophisticated coding tools (and links to other kinds of sources) that NVivo offers. Transana makes transcription a breeze, but it would be useful to be able to code as I go along with the transcription process, instead of waiting until the transcription is done and then converting it to NVivo for coding.

 

Ann

Link to post
Share on other sites
  • 2 weeks later...

Hi Ann, Alan and Jacob

 

Just wanted to thank you for your feedback - we are looking at ways to improve the transcription experience in NVivo and it's great to hear from NVivo users on the coal face!

 

Cheers,

Kath

Link to post
Share on other sites
  • 4 months later...

Hi there,

Thank you so much for this information - I was despairing of ever being able to code comprehensively in transana, but Nvivo is too expensive for me to get a copy for my transcriber. However, I'm having problems importing the transcripts as described - my time stamps are not being recognised (into Nvivo 8). I've pasted an extract of my transcript, which is in a .rtf doc, in case that helps identify the problem.

 

Help greatly appreciated.

 

Deb

 

 

 

0:00:06.2 ¤<6160> *****Melen's group playing Reef Game. Audio is either absent or spotty, until ¤<499040> *

0:08:19.0 ***.

0:08:29.2 ***Bakit walang kita ang factory? Ba't nawala? Why hasn't the factory made money? Why is it gone?

0:08:32.7 Unseen: ¤<512736> *****Wala. Kapag ** ***isda, Nothing. When the fish are ** ***Unseen: ¤<515971> **

0:08:36.0 ***Isa lang kumpanya nalugi.Only one company went under.Sam: ¤<519054> **

0:08:39.1 ***Hindi man lang kumita yung fish **beams ***niya?Its fish **beams ***didn't even make money? UnseenW: ¤<522482> **

0:08:42.5 ***Sorry.Unseen: ¤<522974> **

0:08:43.0 ***Ang fish cage, walang kita? The fish cage didn't make money? UnseenW: ¤<524020> **

0:08:44.0 ***Wala kasi ano daw.None, because it is Unseen: ¤<526215> **

0:08:46.2 *****Agpas ***Unseen: ¤<527145> **

0:08:47.1 ***Malas. Hindi **nag-- ***Bad luck. It didn't ** ***Unseen: ¤<529430> **

0:08:49.4 ***Ay, aalis muna ako ng fish cage, papalaot na rin ang aking anak.Oh, I'll leave the fish cage, my kid is going out to sea.sMelen: ¤<533524> **

0:08:53.5 ***Di itong Player 2 ang na, ang walang kita, no? 'Wawa naman. So yung

Link to post
Share on other sites

Hi Deb,

 

I copied your transcript into a Word doc, saved it as RTF and successfully imported it into NVivo - on import, did you choose the Timestamp option?

 

forum_transcripts.png

 

Here is the RTF file that I made:

 

For forum_Transcripts.rtf

 

You could try importing it using these steps:

  1. In the audio/video source, click the yellow bar to switch to Edit mode.
  2. On the External Data tab, in the Import group, click Rows.
  3. Browse and select the rtf file.
  4. Choose Timestamp from the Create one row for each list.
  5. Click OK.

 

Let me know how you get on,

 

Cheers,

Kath

Link to post
Share on other sites
  • 1 month later...

Dear Kath (and David)

Thanks very much - is it possible it was a version issue? I was using 8, but since seeing your screendump, after the lovely David Tran fixed up my user problems, I followed the same instructions on version of 9, and it worked.

 

Sorry about the delay - it seems there was a problem with email notification, which has hopefully been fixed

Cheers

deb

  • Like 1
Link to post
Share on other sites
  • 7 months later...

I would like to add three four requests for improving the use of transcripts in NVivo:

  1. (This seems to be a bug) When the transcript table is in focus and I use the mouse wheel, I guess the table should scroll. It sometimes does scroll, but rather unpredictably so. (Update: the problem seems to be that scrolling does not work when the cursor is blinking inside the transcript table. It only works when you highlight a whole row, which is pretty counterintuitive and inconvenient)
  2. The scrolling should be "softer". Right now, if I move the mouse wheel one nudge further, the transcript table moves a whole page, which leaves me without orientation where I am and how far I have moved. One stop on the mouse mousewheel should equal one row, not one page (or whatever it is now).
  3. NVivo should have an import function to import transcripts produced with f4 (http://www.audiotran.../english/f4.htm)
  4. Scrolling does not even work properly when I try to do it "manually", i.e. when I hold the scroll bar on the right with the mouse and move it up and down nothing happens until I release it. That makes this function pretty useless, since I don't know when to release the mouse button unless I see where in the transcript I am...

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
×
×
  • Create New...