How To Track Multiple Speakers
Most audio files only have one or two speakers.
However, there are some files, like group interviews, panels, or focus groups, where you’ll have to assign speaker tags and identify more than two speakers. That’s why learning how to track speakers is vital to transcriptionists.
Tracking speakers isn’t easy. It’s one of the most difficult tasks for transcriptionists. But if you can track speakers well, you have the opportunity to earn more money.
Files with multiple speakers pay more than single-speaker or two-speaker files. The key to tracking speakers is being able to recognize different voices.
Usually, when you receive a file, your company will tell you how many speakers there are in the audio.
Sometimes you’re lucky and speakers will identify themselves in the beginning so that you can place a voice with a name right away.
Other times, you’ll have to figure it out as you go. The most helpful tool to help you track speakers is the Bookmarks tool in Express Scribe.
Bookmark feature is very useful for audios with multiple speakers.
It can be very difficult to identify speakers if there are more than two people speaking.
So sometimes you have to go back and re-listen to their voices to identify who is speaking.
You can use the bookmark feature to help you do this.
Let’s say, for example, you have four speakers:
Speaker 1 speaks at 0:00:00.
Speaker 2 speaks at 0:04:00.
Speaker 3 speaks at 0:08:00.
Speaker 4 speaks at 0:10:45.
You can bookmark each speaker for your reference later in the audio if you’re having trouble identifying who is speaking.
To set a Bookmark, click Bookmark at the desired time.
To choose a Bookmark, click Bookmark> Open Bookmark List.
Select a bookmark.
- Multiple-Speaker-Template
- Marketing Panel Discussion Partial Transcription Answer
- Marketing-Panel-Discussion-Partial-AudioS
- Marketing Panel Discussion Full Transcript Answer
- Marketing-Panel-Discussion-Full-Audio
There’s two versions ohttps://www.typewhizz.co.uk/wp-content/uploads/2021/10/Lecture%2024%20-%20Practice%20File%20Marketing%20Panel%20Discussion%20-%20Instructions.zipf this file. One is long and will give you lots of practice transcribing. The other is a shortened version, which has the beginning, where the panel introduces themselves, and then the Q&A at the end.
Although this file has lots of speakers, the speakers introduced themselves at the beginning, which makes it a bit easier.
We will label the speakers as Moderator, the person running the discussion, and then use the first name of each speaker. Since the speakers use their full names, as well as company titles, you can research the names’ spelling as well as the company’s spellings.
So let’s go ahead and start the audio and type out the first few tags. At this point, we are not going to add any bookmarks.
Moderator: So, if we could, we’ll just have Brian start off here.
Brian: Brian Phillips, Executive Creative Director and CoFounder of the Basement.
Jeb: Jeb Banner, CEO of SmallBox and Founder, and, as of today, Chair of The Speak Easy.
Gail: Hi, I’m Gail McDaniel. I am Global Marketing Manager for the Aortic Business Unit at Cook Medical.
Now that the first few speakers are labeled, we will start the audio over again and put in a bookmark for each speaker besides the moderator.
We are not bookmarking the moderator for this file because we can easily tell who the moderator is by their questions.
In other files, you may want to label the moderator as well.
Brian: Brian Phillips, Executive Creative Director and CoFounder of the Basement.
As Brian introduces himself, we are going to go up to the Bookmarks menu and select Set Bookmark.
For the description, we will put Brian.
This puts a bookmark in the audio right where Brian’s speaking, along with his name. That way, we can come back to it later.
Let’s go through and add bookmarks for Jeb and Gail.
Jeb: Jeb Banner, CEO of SmallBox and Founder, and, as of today, Chair of The Speak Easy. Gail: Hi, I’m Gail McDaniel. I am…
For most of our files, we add the bookmarks at the same time I’m transcribing, instead of going back and adding bookmarks.
But we thought it would be easier to show you how to label speakers and add bookmarks this way.
Gail takes a while introducing herself, and so we will fill in her introduction. The next speaker, George, comes in at the two-minute mark, so let’s type out his introduction and add a bookmark for him.
George: I’m just George Evans, Founder, Principal, and Creative Director at Brandwidth.
Finally, we will add a label and bookmark to the last speaker, Doug.
Doug: Doug Karr. I think some of you guys know me. I run the Marketing Tech Blog and have an agency, DK New Media. We do search and social media marketing for our clients. And then, we just launched an email marketing company called CircuPress.
Now, we have all of our speakers labeled.
Gail’s voice will be the easiest to track since she’s the only female, and so her voice is very distinct. It will take more effort to track the other speakers’ voices, although some are more unique than others.
Let’s go ahead and listen to all of their voices, so we can see how they’re different.
Recording:
Jeb Banner, CEO of SmallBox and Founder and President… Global Marketing Manager for the Aortic… Founder, Principal, and Creative Director at Bandwidth. I run the Marketing Tech Blog and have an agency, DK New Media. We do search and social…
Let’s go ahead and paste in some notes I wrote earlier about their voices. After the speakers introduce themselves, the moderator asks a long question.
Let us jump down further in the file, at 00:03:36, when the first speaker who is not the moderator starts.
So, we are going to listen to this voice and bookmark this section, as well.
Express Scribe won’t save your current place when you check bookmarks.
So, we usually save a bookmark of our current place and then rename it when we’ve figured out who the speaker is.
Another option is to copy your audio file and to use one file to transcribe the audio and the other file to put in your bookmarks since Express Scribe will keep your place for both files.
Recording: Most of the companies that come to us, they know they’re broken. So, they know that they have…
As we are listening, you may realize that it’s not too low. It’s more of a medium tone, so we’ll save our current place and go back through to bookmarks and find the one that matches.
Recording: Jeb Banner. CEO of SmallBox, Founder, Principal, and Creative I run the Marketing Tech blog. So, they know that they have a website and they have a web presence, but it’s not bringing in…
Now that you listened to all the files, the one that it sounded most similar to was Doug. And so, we are going to say that that’s a match and label the speaker as Doug.
If we are not sure of a speaker, we put a question mark next to their name in the bookmark at the current section.
So, if I wasn’t 100% sure this was Doug, I would put a question mark next to the voice and hope that, later in the file, I would be able to match this voice to Doug’s voice further on. I can also put brackets around the speaker’s name to indicate that my guess is that it’s this speaker, but I’m not positive.
It’s similar to how you put brackets around words that you’re unsure of. Sometimes, the context of what is being said, such as a company name or other speakers referencing the previous speaker by name, will give me clues to who a speaker is.
When speakers talk for only a few words, it can be hard to identify them. That’s when I’m more likely to guess and put their name in brackets. Sometimes, I’ll even just label it as ‘male’ with brackets, if I don’t have a guess.
As I work my way through the audio, I do my best to figure out the speakers on the first pass. When I proof the file to audio, I get one more chance to label the speakers correctly. Let’s do another speaker.
After Doug talks, another male talks.
So again, I save my spot in this file and listen to all of the voices.
Recording: My guess is you’re going to get a lot of alignment from all… [Inaudible] My guess is, you’re going to get a lot of alignment…
This voice matches George. So, I’ll label this speaker as George and rename the bookmark to George too. That way, I have a longer sample of audio to compare George’s voice against later in the file.
I’m going to fill in more of this file. As we work through this file, we need to label and check speakers’ voices as needed, until we are more familiar with the voices.
You can see that, as we work through the file, our list of bookmarks grows.
One thing to talk about is crosstalk. Handling crosstalk when there’s multiple speakers can be messy.
In the transcript can see an example of some crosstalk that comes up between Gail and Doug. I just use dashes to show that Doug was interrupted and then continued his remarks after Gail. And further on you can see that there were some remarks that were inaudible due to crosstalk.
Tracking speakers is a skill that is developed with practice. If you can master it, you can make more money by taking on harder transcripts. The key is to notice specific characteristics about voices and bookmark them in Express Scribe so that you can refer to them as you transcribe.