AI Transcription

Now, Transistor offers an AI transcriptions add-on. This new feature will automatically transcribe your podcast audio, detect different speakers, and generate outputs in SRT, VTT, JSON, HTML, and TXT formats. We'll also publish the transcript as a web page for each episode, making it even easier for your audience to access and engage with your content. Transcription is available for over 50 languages.

You can also upload transcriptions you've created or downloaded from another source. See how to upload a transcript here.

FAQ

Can I test out AI transcription in my account?

Yes! You will have received 60 free trial credits in your account that you can use to test out our AI Transcription feature on up to 1 hour of audio.

In your Episodes list, click on the grey transcript icon for an episode that doesn't yet have a transcript. That will take you to the Transcripts tab for that episode.



Click the green Transcribe with AI button to start the transcription process.


Once you have used your free trial credits you can purchase a paid add-on to your subscription to have transcription available for more episodes.

How do I enable AI transcription in my Transistor account?

You can now enable AI transcription in your account as a paid add-on. Go to your account page and in the Billing & Subscription section you can enable one of our two plans:

$5 per month for 300 minutes per month
$20 per month for 1200 minutes per month



This provides you with up to 5 or 20 hours worth of AI transcription each month. Your transcription credits will be topped up on your usual monthly or annual renewal date.

Interested in having your back catalogue of episodes transcribed? Reach out to support@transistor.fm or our Live Chat and we can calculate a price for you.

How does AI transcription work?

AI transcription leverages machine learning algorithms and natural language processing technologies to convert spoken language (audio) into written text.

You may have used speech-to-text that leverages a similar technology when you've uploaded a video to YouTube or Facebook, and its automatically generated captions. Google Docs also has a similar speech-to-text feature.

How accurate are AI transcriptions?

AI transcription is significantly cheaper than getting human to manually transcribe text, however it's not 100% accurate. You will need to go over the text that's generated, and manually fix errors in the editor.

Can I correct and edit the AI transcripts?

Yes! Our transcript editor allows you to easily correct words, paragraphs and speaker names. You can learn how here:
How to edit and correct an AI transcript?

Why does my podcast need a transcript?

Transcripts are no longer just a nice-to-have feature; they're a must-have for any serious podcaster. Not only do they help with SEO they also help your podcast be more accessible. It also makes your content easier to search and repurpose. With Transistor's support and new AI transcription feature, adding transcripts to your podcast episodes has never been easier.

What languages can be transcribed?

More than 50 spoken languages are supported. Check out the entire list.

What transcription formats does Transistor provide?

We output transcripts in the five most common formats. They are SRT, VTT, JSON, HTML and TXT formats.

SRT - SubRip Subtitle format is a universal plain text timestamped subtitle format. Used on many platforms like YouTube

VTT - Web Video Text to Track is a more complex subtitle format that can contain metadata like color, font and placement of subtitles

JSON - A precisely timestamped transcript using milliseconds. Each caption is a block with a speaker label and a start and end timestamp. A widely used format that is not human-readable but useful for developers working with transcripts in an application

HTML - A transcript with timestamps that is formatted ready for use on a web page with HTML paragraph and citation tags

TXT - A plain text transcript without timestamps and minimal formatting. Commonly used like a blog post.

Where will the transcripts appear?

We provide you with a hosted transcript web page for each episode. A link to the transcript page is included in your RSS feed so that any apps that support transcripts can display it too.

This is a sample podcast transcript page:

https://share.transistor.fm/s/11a408d1/transcript

Can I still upload my own transcripts?

Yes, you can still upload your own transcripts to episodes. You can find our guide on that here:
How to add transcripts to your podcast episodes

Where can I ask other questions about AI Transcripts?

If you have any questions about our new feature please reach out to us at support@transistor.fm or on our Live Chat widget here: https://transistor.fm/contact/

Updated on: 29/02/2024