Skip to main content
Question

Automatically Transcribe Audio files using AI

  • April 24, 2025
  • 2 replies
  • 193 views

Forum|alt.badge.img+1

Hi there! I have created a form in Filout and I am allowing partners to upload their answers using an audio file. This is automatically linked to AirTable. I am not trying to use the Airtable AI Function to automatically create transcripts. Whenever I try to run it, I just receive “Unknown Error, please try again”

 

Any help with how to automatically transcribe my audio responses? 

 

Here is what I have in the AI field: 

 

You are a transcription specialist with expertise in converting audio recordings into accurate text transcriptions. Your task is to listen to the provided audio file and transcribe the spoken content into written text.

 

Begin by carefully listening to the entire audio recording, ensuring that you capture every word and nuance. Pay attention to the speaker's tone, context, and any specific terminology used. Do not use the internet to find additional information; rely solely on the audio content provided.

 

Output the transcription as plain text, without any additional formatting or headings. If you cannot transcribe the audio, output "Transcription not possible." and explain why. Ensure that the transcription is clear and concise, capturing the essence of the spoken words. For example, if the audio says "Hello, welcome to our event," the output should be "Hello, welcome to our event."

 

Context and Data:

Audio File:

 

Output:

2 replies

RuddConsulting
Forum|alt.badge.img+10
  • Participating Frequently
  • April 24, 2025

I don’t think it can read .mp4 files currently. ​@Mikayla_LaRosa do you know?


ScottWorld
Forum|alt.badge.img+35
  • Genius
  • April 24, 2025

​Hi @CPTech,

Airtable’s AI does not currently support audio files.

Your best bet would be to trigger an automation with Make’s advanced automations and integrations, where you can use OpenAI or Perplexity or Gemini or any other AI tool that may support transcription of audio files.

After the transcription is done, your automation would send the transcription text back to Airtable.

However, even better, instead of using AI to transcribe your audio files, you can use a professional transcription service such as 3Scribe Transcriptions and then use Make’s 3Scribe integrations to do the professional transcription for you.

In fact, you can use any transcription service that you’d like, as long as they have an API that you can tap into.

Even if Make doesn’t natively support a particular transcription service, you can always write your own custom API calls with Make’s HTTP module.

If you’ve never used Make before, I’ve assembled a bunch of Make training resources in this thread. For example, here is one of the ways that you could instantly trigger a Make automation from Airtable.

I also give live demonstrations of how to use Make in many of my Airtable podcast appearances. For example, in this video, I show how to work with Airtable arrays in Make. The reason that learning about arrays might be important to you is because attachment fields are arrays.

Hope this helps! If you’d like to hire the best Airtable consultant to help you with anything Airtable-related, please feel free to contact me through my website: Airtable consultant — ScottWorld