How to transcribe an interview in 5 simple steps (2024)

If you want to learn more about different transcription options, here’s what you need to know.
June 13, 2023
Brandon Copple
In this article
Start editing audio & video
This makes the editing process so much faster. I wish I knew about Descript a year ago.
Matt D., Copywriter
Sign up

What type of content do you primarily create?

Videos
Podcasts
Social media clips
Transcriptions
Start editing audio & video
This makes the editing process so much faster. I wish I knew about Descript a year ago.
Matt D., Copywriter
Sign up

What type of content do you primarily create?

Videos
Podcasts
Social media clips
Transcriptions

So you’ve finished an interview recording

Whether you’re using that conversation in an article or a research project or editing it into a video or podcast, it’s time to transcribe it. 

Where do you start? You start by typing “how to transcribe an interview” into Google, but you probably already did, and that’s what brought you here. Hi!

Anyway, you’re in the right place, because Descript will automatically transcribe your interview (and let you edit the audio or video right from the transcript). But if you want to learn more about different transcription options, here’s what you need to know.

What is an interview transcript?

An interview transcript is an audio track converted into text so people can read it. Before you start transcribing, it’s worth considering the transcript’s ultimate purpose.

  • Are you using the interview as a research source? If you just want to drain your subject’s brain with interview questions, and their exact quotes don’t matter, you don’t have to waste your time with an intelligent verbatim transcription. A quick and dirty job will do.
  • Will you be quoting your subject for publication? People don’t typically like being misquoted. Yet, it rarely reads well when people are quoted exactly as they speak in real life. You’ll likely want to split the difference: Go through that audio with a fine-tooth comb, get it down as faithfully as you can, and eliminate all the “ums,” “ahs,” and unnecessary “I means.” No one will fault you for making them sound more articulate.
  • Is every single word, pause, and false start important? You’ll have to do a full verbatim transcript. When details count, they really count, so having every detail from your interview will be helpful for your project.

How to transcribe an interview, the easy way

1. Get transcription software

Praise the robots, for they allow us to save time. 

Not only is a lot of transcription software free (or free to a point), but it also takes the drudgery out of the transcription process. Some errors and typos are inevitable, so you’ll want to edit the text afterward, even if you use a high-quality transcriber like Descript.

Descript automatically transcribes any file you import; just drag your audio into the app, wait a few seconds, and you’ll have a complete transcript. 

That, in our very biased opinion, makes Descript the best transcription service, hands down. But here are some other transcription tools:

  • Otter. Otter is a workplace tool to transcribe meetings and follow up with any actionable items easily. But you could also just as easily use it to transcribe your interviews in real time. Otter uses AI to automatically separate speakers and transcribe as your interview happens, and the basic features are free. You'll have to upgrade to get more advanced functions, like importing files for transcription. Plans start at $8.33/month.
  • Amberscript: Amberscript, another AI-based transcription software, also offers translation and human editing services. The downside is that you only get 10 minutes of automatic transcription free. Beyond that, you’ll need to pay $10 per hour of audio.
  • Scribie: For automatic transcription, Scribie claims to reach over 80% accuracy and only costs 10 cents per minute. That’s definitely good for the price point, though for the interviewer on the go, it’s worth noting that the service doesn’t have a mobile app. 

2. Upload your audio or video file

Open your Descript app, then click New project in the upper right corner.

On the following screen, name your project and click Choose a file to transcribe.

Choose your file.

After selecting open, Descript will automatically transcribe your audio or video file. 

3. Add speakers

Next, Descript will ask you to identify the speakers in your file. If it’s just you and another interviewee, select two from the dropdown menu.

Descript will then ask you to identify each speaker. We'll play you a short clip from your file, and you'll type in the speaker's name. 

Then click Add “Name” as speaker. Easy peasy!

4. Clarify the transcript as needed

Now you’ve got a transcript ready for editing. 

You probably see some mistakes already. With Descript, you can edit your audio by editing the text in the transcript. 

If you don’t like a word or sentence in your transcript, highlight it and press delete on your keyboard. You can also correct words and punctuation quickly with a few handy keyboard shortcuts

You can also choose to automatically correct any mistakes the AI made. Access them in the upper right hand corner of your transcript.

You’ll find shortcuts like:

  • Shorten word gaps. Remove long periods of silence from your recording automatically. Set how many seconds (or more) of gaps to fill, then reduce them to whatever you want. 
  • Remove filler words. Descript will automatically remove filler words like "you know," "well," or “um”. It can also remove stutters and repetitions. 
  • Detect transcription errors. Clean up your transcript quickly with this tool. Click this button and Descript will highlight probable recording errors for your review.

5. Export the transcript and proofread

At this point, you’ve got a clean transcript ready to go. The last step here is to export the transcript and give it a good proofread. 

First, click the Publish button in the top right corner of your screen. Click the Export tab and then the Transcript button

Before exporting the file, you can choose to include:

  • Composition name
  • Markers
  • Ignored text
  • Speaker labels

You can also choose how often you want the timestamps to appear in your transcript file, and the format of the transcript file. Once you hit Export, you can find the file on your computer. 

Here’s an example of what we just made.

💡PRO TIP: If you’re publishing the transcript as-is, hire an editor or use Grammarly to tighten it up. There will likely be grammar errors you missed during editing. It’s normal. Edit the transcript before pushing it live to truly make it shine out in the wild. 

Other ways to transcribe an interview

The above process for transcribing interviews is super easy, but if you’re up for a bit more of a challenge, or have more budget, there are two other types of transcription methods. 

Transcribe manually

Transcribing interviews manually is the “roll up your sleeves and type like the wind” approach. There aren’t many fancy tools here, just you, your audio recording, some headphones, and a word processor. (There are tools that will let you slow down and go back in a recording without leaving the keyboard, though — our managing editor has trusted the manual transcription tool from Wreally for years.)

The upsides are that it’s free (assuming you have the basic interview recording equipment), and you’ll get very familiar with the material. The downsides are that it is time-consuming, and if you don’t have fast typing speed, it could feel like a Sisyphean task.

Here’s a rundown of how to get started.

  1. Gather your materials. You’ll want a pair of headphones, a decent word processor, and a keyboard.
  2. Listen to the entire interview first. If possible, you should listen to the interview soon after you do it so it’s fresh in your mind. If you or your subject dropped a word or mumbled, you might remember what was said within a day or two. But if it’s been a week or more since the interview, you probably won’t.
  3. Do a rough pass. Perfectionists might struggle with this step, but listen to the audio track and do your best to type everything out without stopping. This will give you an editing basis. It will probably look like a drunk person’s manifesto when you’re done, but that’s fine.
  4. Edit your rough pass. Before you listen through again, fix up your messy transcript the best you can. Separate the speakers from one another and correct any words or phrases you can.
  5. Listen again and adjust your transcript. Now that you have a serviceable document, follow along with your transcript as you listen again, tweaking words or phrases as necessary. If you’re lucky and your subject spoke slowly, this may be the only extra step you need.
  6. Listen through and adjust again as needed. As mentioned earlier, if you only need the transcript for the information you got from the subject, you likely don’t need to worry about getting everything exactly right. If you do need precise quotes, listen to your recording and edit your transcript until the two match up.

Hire a transcription service 

Nothing beats that human touch — for now at least. (The robots will replace all of us in the end.) Sending your audio files to a service that will farm it out to an actual person for transcription has obvious benefits. 

Namely, people understand the nuances of human speech better than a computer will, and if you hire a quality service, you won’t have to worry too much about editing what they send back. 

On the other hand, a transcript made by a human will cost you more than one made by software, and the level of detail means it’ll also take a lot of time.

Here are a handful of respected agencies to try:

  • Descript White Glove: You had to know this one would be in here. It’s true: Descript allows you the option of ordering a human transcript right from the app with its White Glove service. It only extends to English for now and there’s a cap of two hours per transcription, but provided the audio quality is decent, Descript will return a transcript that’s up to 99% accurate.
  • Vanan Services: Vanan provides transcription services in a variety of languages and also offers a free trial of up to five minutes from a given file. If it matters to you (and it should, because it’s cool), even NASA has used their transcription services. Their exact rate depends on the audio, but they do provide quotes in advance.
  • GoTranscript: For a basic file with one or two speakers, GoTranscript’s rates start at 90 cents a minute, and they deliver reliable results. Everything beyond that (full verbatim transcription, timecoding, etc.) will cost you extra per minute.

Example of interview transcript 

Here’s an example of an edited transcription with timestamp headers: 

Interviewer (Content Marketer, Susan): Good afternoon, Mike. Thank you for taking the time to be with us today.

Mike (Founder, MarketingTool.io): Hi Susan, thank you for having me.

[00:00:15]

Susan: Mike, can you briefly tell us what inspired you to create MarketingTool.io?

Mike: Absolutely. I saw a gap in the market for user-friendly and cost-effective marketing tools. I wanted to create something that every business could use.

[00:00:35]

Susan: Interesting. What's the core functionality that sets MarketingTool.io apart from other tools?

Mike: Our AI-driven analytics. It not only provides accurate data but also insightful suggestions on how to improve your marketing strategies.

[00:01:00]

Susan: That sounds like a game-changer. Can you tell us about a significant challenge you faced while developing this tool?

Mike: Building the AI was the toughest part. It had to be smart yet straightforward. The balance was challenging but we managed it.

[00:01:30]

Susan: I can imagine that was a significant task. Moving forward, what new features can your users expect from MarketingTool.io?

Mike: We're planning to integrate more social media platforms and add real-time tracking capabilities.

Final thoughts

The nice (and maddening) thing about transcribing is that you know you can always get a free transcription by just doing it yourself. 

But if you have the option of hiring a transcriptionist to do it, it’s definitely something to consider. Transcribing an interview podcast is fun once or twice, but if you have a whole bunch of long, say, qualitative research interviews to get into text, you’ll be kicking yourself and/or going insane before long.

Interview transcription FAQ

What is the easiest way to transcribe an interview?

The easiest way to transcribe an interview is by using an automated transcription service like Descript. These apps use AI to convert spoken words into text, reducing the manual effort and time required to transcribe an interview.

How would you transcribe an interview?

To transcribe an interview, you would simply upload the audio or video file to a software like Descript. It’ll generate a written transcript in minutes and you can edit for accuracy. 

How long does it take to transcribe a 60 minute interview?

If you’re using an audio-to-text app, the turnaround time to transcribe a 60 minute interview is minutes. You’ll need to spend some time reviewing and correcting errors. If you’re transcribing the interview manually, it could take you up to six hours considering you have to pause, rewind, and review way more often.

Brandon Copple
Head of Content at Descript. Former Editor at Groupon, Chicago Sun-Times, and a bunch of other places. Dad. Book reader. Friend to many Matts.
Start creating
The all-in-one video & podcast editor, easy as a doc.
Sign up
Start creating—for free
Sign up
Join millions of others creating with Descript

How to transcribe an interview in 5 simple steps (2024)

Blue computer monitor and blue podcast microphone on a blue background

So you’ve finished an interview recording

Whether you’re using that conversation in an article or a research project or editing it into a video or podcast, it’s time to transcribe it. 

Where do you start? You start by typing “how to transcribe an interview” into Google, but you probably already did, and that’s what brought you here. Hi!

Anyway, you’re in the right place, because Descript will automatically transcribe your interview (and let you edit the audio or video right from the transcript). But if you want to learn more about different transcription options, here’s what you need to know.

Transcribe. Edit. As easy as tapping your backspace key.
Create your podcast from start to finish with Descript.

What is an interview transcript?

An interview transcript is an audio track converted into text so people can read it. Before you start transcribing, it’s worth considering the transcript’s ultimate purpose.

  • Are you using the interview as a research source? If you just want to drain your subject’s brain with interview questions, and their exact quotes don’t matter, you don’t have to waste your time with an intelligent verbatim transcription. A quick and dirty job will do.
  • Will you be quoting your subject for publication? People don’t typically like being misquoted. Yet, it rarely reads well when people are quoted exactly as they speak in real life. You’ll likely want to split the difference: Go through that audio with a fine-tooth comb, get it down as faithfully as you can, and eliminate all the “ums,” “ahs,” and unnecessary “I means.” No one will fault you for making them sound more articulate.
  • Is every single word, pause, and false start important? You’ll have to do a full verbatim transcript. When details count, they really count, so having every detail from your interview will be helpful for your project.

How to transcribe an interview, the easy way

1. Get transcription software

Praise the robots, for they allow us to save time. 

Not only is a lot of transcription software free (or free to a point), but it also takes the drudgery out of the transcription process. Some errors and typos are inevitable, so you’ll want to edit the text afterward, even if you use a high-quality transcriber like Descript.

Descript automatically transcribes any file you import; just drag your audio into the app, wait a few seconds, and you’ll have a complete transcript. 

That, in our very biased opinion, makes Descript the best transcription service, hands down. But here are some other transcription tools:

  • Otter. Otter is a workplace tool to transcribe meetings and follow up with any actionable items easily. But you could also just as easily use it to transcribe your interviews in real time. Otter uses AI to automatically separate speakers and transcribe as your interview happens, and the basic features are free. You'll have to upgrade to get more advanced functions, like importing files for transcription. Plans start at $8.33/month.
  • Amberscript: Amberscript, another AI-based transcription software, also offers translation and human editing services. The downside is that you only get 10 minutes of automatic transcription free. Beyond that, you’ll need to pay $10 per hour of audio.
  • Scribie: For automatic transcription, Scribie claims to reach over 80% accuracy and only costs 10 cents per minute. That’s definitely good for the price point, though for the interviewer on the go, it’s worth noting that the service doesn’t have a mobile app. 

2. Upload your audio or video file

Open your Descript app, then click New project in the upper right corner.

On the following screen, name your project and click Choose a file to transcribe.

Choose your file.

After selecting open, Descript will automatically transcribe your audio or video file. 

3. Add speakers

Next, Descript will ask you to identify the speakers in your file. If it’s just you and another interviewee, select two from the dropdown menu.

Descript will then ask you to identify each speaker. We'll play you a short clip from your file, and you'll type in the speaker's name. 

Then click Add “Name” as speaker. Easy peasy!

4. Clarify the transcript as needed

Now you’ve got a transcript ready for editing. 

You probably see some mistakes already. With Descript, you can edit your audio by editing the text in the transcript. 

If you don’t like a word or sentence in your transcript, highlight it and press delete on your keyboard. You can also correct words and punctuation quickly with a few handy keyboard shortcuts

You can also choose to automatically correct any mistakes the AI made. Access them in the upper right hand corner of your transcript.

You’ll find shortcuts like:

  • Shorten word gaps. Remove long periods of silence from your recording automatically. Set how many seconds (or more) of gaps to fill, then reduce them to whatever you want. 
  • Remove filler words. Descript will automatically remove filler words like "you know," "well," or “um”. It can also remove stutters and repetitions. 
  • Detect transcription errors. Clean up your transcript quickly with this tool. Click this button and Descript will highlight probable recording errors for your review.

5. Export the transcript and proofread

At this point, you’ve got a clean transcript ready to go. The last step here is to export the transcript and give it a good proofread. 

First, click the Publish button in the top right corner of your screen. Click the Export tab and then the Transcript button

Before exporting the file, you can choose to include:

  • Composition name
  • Markers
  • Ignored text
  • Speaker labels

You can also choose how often you want the timestamps to appear in your transcript file, and the format of the transcript file. Once you hit Export, you can find the file on your computer. 

Here’s an example of what we just made.

💡PRO TIP: If you’re publishing the transcript as-is, hire an editor or use Grammarly to tighten it up. There will likely be grammar errors you missed during editing. It’s normal. Edit the transcript before pushing it live to truly make it shine out in the wild. 

Other ways to transcribe an interview

The above process for transcribing interviews is super easy, but if you’re up for a bit more of a challenge, or have more budget, there are two other types of transcription methods. 

Transcribe manually

Transcribing interviews manually is the “roll up your sleeves and type like the wind” approach. There aren’t many fancy tools here, just you, your audio recording, some headphones, and a word processor. (There are tools that will let you slow down and go back in a recording without leaving the keyboard, though — our managing editor has trusted the manual transcription tool from Wreally for years.)

The upsides are that it’s free (assuming you have the basic interview recording equipment), and you’ll get very familiar with the material. The downsides are that it is time-consuming, and if you don’t have fast typing speed, it could feel like a Sisyphean task.

Here’s a rundown of how to get started.

  1. Gather your materials. You’ll want a pair of headphones, a decent word processor, and a keyboard.
  2. Listen to the entire interview first. If possible, you should listen to the interview soon after you do it so it’s fresh in your mind. If you or your subject dropped a word or mumbled, you might remember what was said within a day or two. But if it’s been a week or more since the interview, you probably won’t.
  3. Do a rough pass. Perfectionists might struggle with this step, but listen to the audio track and do your best to type everything out without stopping. This will give you an editing basis. It will probably look like a drunk person’s manifesto when you’re done, but that’s fine.
  4. Edit your rough pass. Before you listen through again, fix up your messy transcript the best you can. Separate the speakers from one another and correct any words or phrases you can.
  5. Listen again and adjust your transcript. Now that you have a serviceable document, follow along with your transcript as you listen again, tweaking words or phrases as necessary. If you’re lucky and your subject spoke slowly, this may be the only extra step you need.
  6. Listen through and adjust again as needed. As mentioned earlier, if you only need the transcript for the information you got from the subject, you likely don’t need to worry about getting everything exactly right. If you do need precise quotes, listen to your recording and edit your transcript until the two match up.

Hire a transcription service 

Nothing beats that human touch — for now at least. (The robots will replace all of us in the end.) Sending your audio files to a service that will farm it out to an actual person for transcription has obvious benefits. 

Namely, people understand the nuances of human speech better than a computer will, and if you hire a quality service, you won’t have to worry too much about editing what they send back. 

On the other hand, a transcript made by a human will cost you more than one made by software, and the level of detail means it’ll also take a lot of time.

Here are a handful of respected agencies to try:

  • Descript White Glove: You had to know this one would be in here. It’s true: Descript allows you the option of ordering a human transcript right from the app with its White Glove service. It only extends to English for now and there’s a cap of two hours per transcription, but provided the audio quality is decent, Descript will return a transcript that’s up to 99% accurate.
  • Vanan Services: Vanan provides transcription services in a variety of languages and also offers a free trial of up to five minutes from a given file. If it matters to you (and it should, because it’s cool), even NASA has used their transcription services. Their exact rate depends on the audio, but they do provide quotes in advance.
  • GoTranscript: For a basic file with one or two speakers, GoTranscript’s rates start at 90 cents a minute, and they deliver reliable results. Everything beyond that (full verbatim transcription, timecoding, etc.) will cost you extra per minute.

Example of interview transcript 

Here’s an example of an edited transcription with timestamp headers: 

Interviewer (Content Marketer, Susan): Good afternoon, Mike. Thank you for taking the time to be with us today.

Mike (Founder, MarketingTool.io): Hi Susan, thank you for having me.

[00:00:15]

Susan: Mike, can you briefly tell us what inspired you to create MarketingTool.io?

Mike: Absolutely. I saw a gap in the market for user-friendly and cost-effective marketing tools. I wanted to create something that every business could use.

[00:00:35]

Susan: Interesting. What's the core functionality that sets MarketingTool.io apart from other tools?

Mike: Our AI-driven analytics. It not only provides accurate data but also insightful suggestions on how to improve your marketing strategies.

[00:01:00]

Susan: That sounds like a game-changer. Can you tell us about a significant challenge you faced while developing this tool?

Mike: Building the AI was the toughest part. It had to be smart yet straightforward. The balance was challenging but we managed it.

[00:01:30]

Susan: I can imagine that was a significant task. Moving forward, what new features can your users expect from MarketingTool.io?

Mike: We're planning to integrate more social media platforms and add real-time tracking capabilities.

Final thoughts

The nice (and maddening) thing about transcribing is that you know you can always get a free transcription by just doing it yourself. 

But if you have the option of hiring a transcriptionist to do it, it’s definitely something to consider. Transcribing an interview podcast is fun once or twice, but if you have a whole bunch of long, say, qualitative research interviews to get into text, you’ll be kicking yourself and/or going insane before long.

Interview transcription FAQ

What is the easiest way to transcribe an interview?

The easiest way to transcribe an interview is by using an automated transcription service like Descript. These apps use AI to convert spoken words into text, reducing the manual effort and time required to transcribe an interview.

How would you transcribe an interview?

To transcribe an interview, you would simply upload the audio or video file to a software like Descript. It’ll generate a written transcript in minutes and you can edit for accuracy. 

How long does it take to transcribe a 60 minute interview?

If you’re using an audio-to-text app, the turnaround time to transcribe a 60 minute interview is minutes. You’ll need to spend some time reviewing and correcting errors. If you’re transcribing the interview manually, it could take you up to six hours considering you have to pause, rewind, and review way more often.

Featured articles:

Video

10 excellent Loom alternatives to record your screen

Looking for the best ways to record your screen without Loom? Find out 10 Loom alternatives, their features, and pricing.

For Business

Social media video marketing: Everything you need to know

Learn how to win over potential customers (and the algorithms) with the right social media video marketing tools, strategy, and optimizations.

Video

10 effective how-to videos to engage your audience

How-to videos are a great way to educate your audience. Whether you’re recording a software demo or training instructions, here’s how to create a how-to video.

Articles you might find interesting

For Business

Instagram stats 2023: Insights to enhance your marketing strategy in 2024

For content creators and marketers, understanding Instagram statistics is crucial for creating effective data-based strategies. By looking at statistics and trends, you can identify your audience’s preferences, narrow down the content formats that work best, and identify the ideal posting times that will generate the highest engagement.

For Business

Essentials for an effective social media marketing plan

Learn how you can create a social media marketing plan that’s purely focused on posts for your own accounts, or can include a strategy for working with influencers.

Podcasting

How to upload your podcast to Apple Podcasts: A guide

Learn how to submit your podcast to the audio streaming service that provides millions of different podcasts and lets you search, download, and curate your collections.

Product Updates

Ultra Fast Audio Synthesis with MelGAN

In this post, we introduce MelGAN, a new generative model of raw audio waveforms created by the Lyrebird team that is capable of generating natural sounding speech at a rate of more than 2,500,000 audio samples per second — more than 100x faster than real time, and 10x faster than alternative methods on similar hardware.We believe that MelGAN paves the way for taking many real-time speech applications onto smaller devices. Imagine, for example, in the not too distant future, having real-time text-to-speech translation on your mobile device without the internet. And, it’s application to music translation brings us one step closer to AI-assisted music composing.We’ve open sourced MelGAN and we encourage interested machine learning developers and researchers to check out our code base.

Podcasting

Podcasting by the numbers: 15 podcast stats for 2024

Who’s driving the growth of the podcasting industry? What do we know about podcast listeners and their behavior? And what can we glean to help podcasters attract those listeners to their show?

For Business

How to Record A Lecture: 7 Tips to Enhance the Learning Experience

The pandemic has forced educators to adopt virtual teaching forms, like record lectures. Luckily, good recordings don’t require much more than software, a webcam and microphone, and a good strategy.

Related articles:

Share this article

Get started for free →