How to Record High-Quality Audio

We are often asked how to improve the accuracy of transcriptions so that getting insights from audio or video recordings can be made even faster. It is unfortunately not so possible to have 100% accurate speech to text conversion today. The good news is that technology is advancing every day and we will soon experience an enhanced performance that will ease your job.

Having said that, based on our and our users’ experience, there are a number of methods you can use to minimize the work necessary in editing documents manually. 

Two women recording an interview

Photo by Kate Oseen on Unsplash

Let’s take a look at our Top 5 Tips to maximize your productivity;

1. Use the recording function of the platform/tool that you are conducting the interview, not another tool.

Some of our users mentioned that they are using their smartphones to record a conversation they are having on other video platforms. This is not a good idea :) If you are having a meeting on Zoom, Skype, Hangouts, or in any other software platform, we highly recommend that you use the recording function of that same platform.

2. Keep the distance to your device short. 

The sound quality of the audio or video increases when you keep a short distance to the device you are talking to. If you are speaking to a phone, it helps if you hold the phone close to your mouth when speaking; if that is not possible, placing the phone just in front of you on the table will also help. If the phone is further away somewhere else, the quality of the audio will decrease.

3. Use headphones and if possible a microphone. 

The headphone prevents echoing of your and your counterparts’ voices. Using headphones help improve the overall sound quality and thus provides you with more free time to analyze the data once the audio or video is automatically transcribed.

We know that not everyone might possess a high-quality microphone when conducting online interviews. Depending on the frequency of your interviews however, a microphone might be worth the investment because it would greatly enhance the sound quality of your interview hence the transcription output. It is not a must-have but if you conduct a lot of interviews and seek the optimal performance on Scrintal, it can be a game-changer.

4. Eliminating the background noise or wind

This might not seem so trivial at first but it does make a huge difference. Elimination or at least minimization of any other conflicting noise in the background helps the software’s algorithm differentiate the sounds more effectively. Imagine a situation in which you were speaking with a friend on the phone who was in a quite noisy area or a windy place. You might recall that it was not as easy to comprehend what she was saying as speaking with her in a quiet environment.

5. The sound quality will definitely become better if the speakers do not interrupt each other but rather wait until they are done with their part.

When two or more people speak at the same time, the voices get intertwined which makes it harder for the algorithm to capture different sounds and differentiate speakers.

Scrintal is an AI-powered SaaS tool that converts audio files in mp3, wav, flac, m4a and aac formats, and video files in mp4 and mov formats to text at a fast speed. It significantly shortens the time it takes to transcribe the recordings and analyze the data on the same platform so that you can spend your limited time on things that matter to you. 

We hope that these tips will simplify your workflow, even more, guarantee the best transcriptions, and minimize manual editing.

Ece Kural's profile picture

Ece Kural

PhD Candidate @ Stockholm University