How to create a podcast transcript


Many podcast hosts focus on their audience, which is great. However, some people cannot listen to podcasts. Use transcripts for people who are hard of hearing or who are learning languages. Provide transcripts to allow them to join the podcast host’s audience.

How to make a transcript

To make a transcript, the podcast host has four options:

  1. Speech-to-text
  2. YouTube
  3. Humanized service
  4. DIY

Each method has advantages and disadvantages. I use each method to create transcripts for the same episode.

standard

I use the same criteria to evaluate which method is best.

Speed ​​is one of the factors used to evaluate options. Speed ​​is the speed of development of the transcript. Next, I used accuracy, which is how many errors in the transcript I need to fix. In the end, I used the price, how much does it cost to create the transcript.

How to deal with transcripts

After making the transcript, the host has a variety of options.

Add to website

Transcripts will increase your chances of finding podcasts in Google search.

People will spend time reading podcast transcripts. This will increase the time on site. This will make podcast sites more popular with Google’s search algorithm.

Blog post

The host can obtain the transcript, add some formatting, and turn the episode transcript into a blog post. Blog posts can be posted in different places for people to read. The blog post can contain a link back to the podcast episode to promote the podcast.

Downloadable

The host can let their audience download the transcript. To make the transcript downloadable, the host must place the transcript on a PDF. Then, create a button with the visitor to download content. The host lost the opportunity to increase the domain authority. But the transcript can be downloaded and on the website at the same time.

benefit

Providing transcripts is an extra job for the host. But there are advantages to providing transcripts.

Search Engine Optimization

When the transcript is placed on the website, Google will index the page. Someone might search for some words used in the transcript. Then, a transcript page may pop up in the search results.

For example, someone might search for “open podcast community podcast.” Then, a page from the Building an Indie Business Podcast website may be displayed. Because in every episode, I will mention open podcast community podcasts.

Expand potential audience

By providing transcripts, presenters are expanding their potential audience. Hosts who provide transcripts have an advantage over those who do not. Those who need transcripts will not become fans of content they cannot consume.

Speech-to-text

The Speech-to-Text software I use to transcribe podcasts is otter.ai. Otter has a free plan that provides users with 600 minutes of free time per month. And 3 free audio files can be uploaded and transcribed, each transcribing can be 45 minutes.

Otter can be used as an audio backup. If there are any problems with regular recording, the host can export the audio from Otter and use it.

The host can export a time-stamped transcript from Otter, just as a monologue turns into a text wall and a txt file. Premium services allow unlimited minutes, upload and recording time. At the same time, audio can be exported as DOCX, PDF and SRT files.

result

I used a 28-minute episode to test the transcript. This episode is about launching a podcast book about Product Hunt.

There are 84 errors in the transcript of the episode. Most of the errors are “Product Hunt” and “Hmm”, and the software mistakes them for “product on” and “arm”. I often encounter “um” problems when recording. The software mistakes it for an “arm”.

It takes 5 to 10 minutes to process the records, and about 45 minutes to fix errors. It takes approximately 1 hour to create and edit transcripts using Speech-to-Text.

Speech to text problem

online

As a service, Otter is a web application. If the host’s internet fails, the transcript will be lost. If the host does not use their three uploads, they can upload the audio file to Otter.

sound

I uploaded one of the interviews I did. I noticed that the service has problems with voice. The service cannot distinguish between the person I interviewed and me. If the host conducts an interview, this is what they need to edit by themselves.

How to use Speech-to-Text software to record

Open the Speech-to-Text software, and then click Record. Speech-to-Text software only records speech. The click of the record button will not be displayed. If you press the record button, then the Speech-to-Text software will open. The sound will appear in the podcast program. The sound of the transcribed recording does not matter. This is because text is the most critical element of Speech-to-Text software.

YouTube

Product Hunt episodes are available 12 hours after the episode is uploaded. There are 31 errors in this episode. Sometimes it will mistake Product Hunt for “product on”. But many times, it understands it, which makes me feel scared.

How to get transcripts from YouTube

There are additional steps to obtaining transcripts from YouTube.

  1. Upload video to YouTube
  2. Go to channel video
  3. Then the details
  4. After that, more options
  5. subtitle
  6. There will be “English” when you are ready
  7. Click on the three dots on “English”
  8. Edit in Classic Studio
  9. action
  10. Download as SRT
  11. Bring the SRT file to another website
  12. Convert SRT to TXT
  13. Fix mistakes
  14. complete

problem

Google

Although YouTube transcripts are free, the host relies heavily on a large company. YouTube has many videos that need to be transcribed. There is no way to know how long the transcript will take to complete. At the same time, they may even think that transcribing the video is not worth it. This happened in a 2-minute video I made. I never got a transcript of the video.

time

YouTube transcripts are accurate and easy to access. Although, the host does not know when they will receive the YouTube transcript. This means that YouTube transcripts are unreliable. The host may record the last episode. They may not be able to provide a transcript of the episode through YouTube.

Humanized service

Podcast hosts can pay for transcription services and have humans write transcriptions for them.

The price of the transcript depends on its length and when the host needs it. For 5-day delivery, they charge USD 0.85/minute, 3-day delivery charges 1 USD/minute, and 1-day delivery charges 1.20 USD/minute. The lowest cost of an episode of Product Hunt is 23.80 US dollars and the highest is 33.60 US dollars.

The service provided me with a free trial for the first 5 minutes. I sent them a two-minute podcast that I made. 3 days later, I sent the Product Hunt episode. I never got any transcripts. Since I have never received a transcript, I cannot evaluate the quality of the work.

The service allows you to choose which option is clean or complete transcript. This means that the host can choose a transcript without all errors, such as “um”. The complete transcript contains all errors. They can also add edits such as timestamps or speaker recognition.

DIY

Podcast hosts can always choose to transcribe episodes themselves. If the podcast is long, this can be a time-consuming task.

In order to try to write my own transcript, I wrote a 1-minute podcast episode. It took 10 minutes to write that minute. This means that a 30-minute podcast episode will take me 5 hours to be fully transcribed. Spending 5 hours on the transcript is not ideal.

The advantage of transcribing this episode yourself is that there are no errors. I can set the style of the transcript in any way.

in conclusion

The best choice for transcription is to use Speech-to-Text software. Once the episode is processed, the host still needs to edit the error. Compared with YouTube or human services, Speech-to-Text software generates more errors. It will take less time to fix the error from Speech-to-Text. Obtaining transcripts from the other two options will take more time. The same principle applies to writing transcripts from scratch. Speech-to-Text software is fast, reliable, and inexpensive.

Some podcast hosts make their viewers pay for transcripts; this is not included. Paying for transcripts is exclusive because podcasts are free. However, they allow people to pay for reading transcripts. this is not right.



Source link

Leave a Reply

Your email address will not be published. Required fields are marked *