Photo by
DALL·E

Proofreading Transcriptions

The Importance of Proofreading Audio Transcriptions

Author
Olga Miroshnyk
Author
Olga Miroshnyk
·
Nov 29, 2022
·
3 min read

It only takes one simple mistake in punctuation or spelling to completely change the structure of your work. If you’re asking your grandma to eat and you forget to add a comma, you’ll end up saying “let’s eat grandma” instead of “let’s eat, grandma.” Commas save lives, and business practices.

When we speak, we don’t think about commas or punctuation- our diction automatically includes it; however, when we utilize audio transcription services- they might not always include those necessary commas and instead have filler words (such as like, oh, um) that can dramatically reduce the quality of our transcription.

That’s when OneAi’s Language Studio Skills come into play. The Proofreading Skill on OneAI takes audio transcriptions and reduces any filler words and speech errors, and for any business it’s incredibly important. First and foremost, poor grammar can damage the credibility of blogs, case studies, how-to guides, and other digital marketing resources - instantly tarnishing SEO and losing potential clientele. 

When we’re taking the effort to create an audio transcription, we care about the results and advantages it’ll bring our businesses. We don’t have to let our hard work go to waste by forgoing one extra step - that’s actually incredibly easy. 

{{cta}}

The Language Studio

First, let’s jump to OneAI’s Language Studio

The studio is in the picture below. The tool input is where you can enter your audio, alongside it is the pipeline box which includes all our Language Skills.

Underneath the pipeline box, you can see the Language Skills, which are different ways we can apply the AI to the content provided. There are different skills we can apply, such as ‘Summarize,’ ‘Proofread,’ ‘Names’, etc. 

proofreading skill One AI

Today we are going to see how “Proofread” works. In today’s example we will use Taylor Swift's NYU's 2022 Commencement speech. This website will help us download Taylor’s speech as an MP3 audio. 

Now, let’s get back to the Language Studio and upload the obtained audio file. Once you do this, you will instantly get the “Transcribe Audio” Language Skill placed into the pipeline box. Add the main hero of our article - the “Proofread” and press “Run Pipeline”. 

Here are our results:

Proofreading Audio Transcriptions

We successfully proofread the audio and generated a new output with 19 replacements made. 

Pay attention also to the yellow notification badge on top of the “+” sign. It hints that you can use additional Language Skills: 

Proofreading Audio Transcriptions One AI

Here are the Language Skills that were detected on top of those we already selected:

Proofreading Audio Transcriptions One AI Studio

Now copy your code and proceed to your code editor.

In the code editor:

Step 1: Pick one of the generated codes you need to your code editor (click here to generate your own API key):

Proofreading Audio Transcriptions One AI API

Step 2: Run “pip install oneai” for Python SDK or “npm install oneai” for Node.js SDK to get the library. Make sure to import all the required packages:

```

import oneai

import base64

oneai.api_key = "[YOUR ONEAI API KEY]"

```

In our example, we used Python SDK code. Here is what we got when we run all the skills together: transcribe and proofread:

```

with open("AudioFile.mp3", "rb") as f:

   

   pipeline = oneai.Pipeline(

     steps = [

           oneai.skills.Transcribe(),

           oneai.skills.Proofread(),

     ]

   )

   transcription = pipeline.run(f)

```

Let’s check our outputs:

Transcription with Proofread:

```

speaker 1: I would like now to introduce Jason king chair and associate Professor of the Clive Davis institute of recorded music., speaker 1: Fish's school of the arts., speaker 1: Who will present the candidate for doctor of fine arts?, speaker 1: Will trustee red rock on, please escort the candidate to the lectern., speaker 2: Taylor Swift., speaker 2: Blazing singer songwriter, producer, director, actress, pioneering and influential advocate for artists rights., speaker 2: And philanthropist., speaker 2: You have brought joy and resolve to your hundreds of millions of fans throughout the world., speaker 2: One of the bestselling music artists in history., speaker 2: You have crossed genres, demographics, age groups, and borders of all kinds., speaker 2: To touch lives around the globe., speaker 2: With 9 original studio albums, two rerecorded studio albums., speaker 2: 5 extended plays., speaker 2: Three live albums and 14 compilations., speaker 2: You have sold well over 100 million album units., speaker 2: Earning awards and honors in every category., speaker 2: You have used the remarkable platform you earned to galvanize support for the equality act to prevent discrimination on the basis of sexual orientation and gender identity., speaker 2: And you have spoken out, and you have supported initiatives to protect women and girls from harassment and sexual assault., speaker 2: You have donated significantly to victims of floods and tornadoes for cancer research., speaker 2: Literacy programs for children and public education., speaker 2: You have fearlessly challenged the exploitation of music artists and successfully championed their right to be compensated for their work., speaker 2: Taylor Swift., speaker 2: You are a role model across the world for your unprecedented talent and accomplishment., speaker 2: Your fierce advocacy for protection of those facing discrimination and your commitment to speaking out forcefully eloquently and effectively on behalf of all artists., speaker 2: By virtue of the authority vested in me, I am pleased to confer upon you the degree of doctor of fine arts on a risk counter., speaker 2: I am now pleased to introduce Taylor Swift, who will respond on behalf of the honorary degree recipient., speaker 1: Hi, I'm Taylor., speaker 1: Last time I was in a stadium this size, I was dancing in heels and wearing a glittery leotard., speaker 1: This outfit is much more comfortable., speaker 1: I would like to say a huge thank you to NYU's chairman of the board of trustees., speaker 1: Bill Berkeley., speaker 1: And all the trustees and members of the board, NYU's president Andrew Hamilton., speaker 1: Provost, Katherine Fleming, and the faculty and alumni here today., speaker 1: Who have made this day possible?, speaker 1: I feel so proud to share this day with my fellow honorees, Susan hawk field, and Felix matzos Rodriguez., speaker 1: Who humble me with the ways they improve our world with their work., speaker 1: As for me, I'm 90% sure the main reason I'm here is because I have a song called 22., speaker 1: And let me just say, I am elated to be here with you today as we celebrate and graduate New York University's class of 2022., speaker 1: Not a single one of us here today has done it alone., speaker 1: We are each a patchwork quilt of those who have loved us., speaker 1: Those who have believed in our futures, those who showed us empathy and kindness., speaker 1: Or told us the truth, even when it wasn't easy to hear., speaker 1: Those who told us we could do it when there was absolutely no proof of that., speaker 1: Someone read stories to you., speaker 1: And taught you to dream., speaker 1: And offered up some moral code of right and wrong for you to try and live by., speaker 1: Someone tried their best to explain every concept in this insanely complex world to the child that was you, as you asked a bazillion questions, how does the moon work and why can we eat salad but not grass?, speaker 1: And maybe they didn't do it perfectly., speaker 1: No one ever can., speaker 1: Maybe they aren't with us anymore., speaker 1: In that case, I hope you'll remember them today., speaker 1: If they are in this stadium, I hope you'll find your own way., speaker 1: To express your gratitude for all the steps and missteps that have led us to this common destination., speaker 1: I know that words are supposed to be my thing, but I will never be able to find the words to thank my mom and dad., speaker 1: My brother Austin, for the sacrifices they made every day., speaker 1: So I could go from singing in coffeehouses to standing up here with you all today because no words would ever be enough., speaker 1: To all the incredible parents, family members, mentors, teachers, allies, friends and loved ones here today who have supported these students in their pursuit of educational enrichment., speaker 1: Let me say to you now, welcome to New York, it's been waiting for you., speaker 1: I'd like to thank NYU for making me technically., speaker 1: On paper, at least., speaker 1: A doctor., speaker 1: Not the type of doctor you would want to round in case of an emergency., speaker 1: Unless you're specific emergency was that you desperately needed to hear a song with a catchy hook and an intensely cathartic bridge section., speaker 1: Or if you're emergency was that you needed a person who can name over 50 breeds of cats in one minute., speaker 1: I never got to have a normal college experience, per se., speaker 1: I went to public high school until tenth grade., speaker 1: And then finished my education doing homeschool work on the floors of airport terminals., speaker 1: Then I went out on the road for radio tour, which sounds incredibly glamorous., speaker 1: But in reality, it consisted of a rental car, motels, and my mom and I pretending to have loud mother daughter fights with each other during boarding, so no one would want the empty seat between us on Southwest., speaker 1: As a kid, I always thought I would go away to college, imagining the posters, I would hang on the wall of my freshman dorm., speaker 1: I even set the ending of my music video from my song love story., speaker 1: At my fantasy imaginary college, where I meet a male model reading a book on the grass, and with one single glance, we realized we had been in love in our past lives., speaker 1: Which is exactly what you guys all experienced at some point in the last four years, right?, speaker 1: But I really can't complain about not having a normal college experience to you because you went to NYU during a global pandemic., speaker 1: Being essentially locked into your dorms and having to do classes over Zoom, everyone in college during normal times, stresses about test scores., speaker 1: But on top of that, you also had to pass a thousand COVID tests., speaker 1: I imagine the idea of a normal college experience was all you wanted to., speaker 1: But in this case, you and I both learned that you don't always get all the things in the bag that you selected from the menu in the delivery surface that is life., speaker 1: You get what you get., speaker 1: And as I would like to say to you, wholeheartedly, you should be very proud of what you've done with it., speaker 1: Today, you leave New York University and then go out into the world, searching what's next., speaker 1: And so will I.. So as a rule, I try not to give anyone unsolicited advice unless they ask for it., speaker 1: I'll go into this more later., speaker 1: I guess I have been officially solicited in this situation to impart whatever wisdom I might have to tell you think that have helped me so far in my life., speaker 1: Please bear in mind that I in no way feel qualified to tell you what to do., speaker 1: You've worked and struggled and sacrificed and studied and dreamed your way here today., speaker 1: And so you know what you're doing., speaker 1: You'll do things differently than I did them., speaker 1: And for different reasons., speaker 1: So I won't tell you what to do because no one likes that., speaker 1: I will, however, give you some life hacks., speaker 1: I wish I knew when I was starting out my dreams of a career., speaker 1: And navigating life, love, pressure, choices, shame, hope, and friendship., speaker 1: The first of which is life can be heavy., speaker 1: Especially if you try to carry it all at once., speaker 1: Part of growing up and moving into new chapters of your life., speaker 1: Is about catch and release., speaker 1: What I mean by that is, knowing what things to keep and what things to release., speaker 1: You can't carry all things., speaker 1: All grudges, all updates on your ex, all enviable promotions, your school bully got at the hedge fund, his uncle started., speaker 1: Decide what is yours to hold?, speaker 1: And let the rest go., speaker 1: Oftentimes, the good things in your life are lighter anyway., speaker 1: So there's more room for them., speaker 1: One toxic relationship can outweigh so many wonderful, simple joys., speaker 1: You get to pick what your life has time and room for., speaker 1: Be discerning., speaker 1: Secondly, learn to live alongside cringe., speaker 1: No matter how hard you try to avoid being cringe, you will look back on your life., speaker 1: And cringe retrospectively., speaker 1: Cringe is unavoidable over a lifetime., speaker 1: Even the term cringe might someday be deemed cringe.,

```

These are the replacements made in the text:

```

print([transcription.transcription.proofread.replacements])

```

```

[[oneai.Label(type='replacement', skill='enhance', output_spans=[Span(start=0, end=6, section=1, text="Fish's")], span_text="Fish's", timestamp=datetime.timedelta(seconds=20, microseconds=750000), value='Tish'), oneai.Label(type='replacement', skill='enhance', output_spans=[Span(start=20, end=24, section=17, text='out,')], span_text='out,', timestamp=datetime.timedelta(seconds=123, microseconds=890000), value='out'), oneai.Label(type='replacement', skill='enhance', output_spans=[Span(start=23, end=32, section=33, text='possible?')], span_text='possible?', timestamp=datetime.timedelta(seconds=310, microseconds=70000), timestamp_end=datetime.timedelta(seconds=310, microseconds=790000), value='possible.'), oneai.Label(type='replacement', skill='enhance', output_spans=[Span(start=65, end=76, section=34, text='hawk field,')], span_text='hawk field,', timestamp=datetime.timedelta(seconds=318, microseconds=50000), value='hawkfield,'), oneai.Label(type='replacement', skill='enhance', output_spans=[Span(start=87, end=93, section=34, text='matzos')], span_text='matzos', value='matos'), oneai.Label(type='replacement', skill='enhance', output_spans=[Span(start=144, end=144, section=46, text='')], timestamp=datetime.timedelta(seconds=417, microseconds=770000), value='like,'), oneai.Label(type='replacement', skill='enhance', output_spans=[Span(start=12, end=12, section=67, text='')], timestamp=datetime.timedelta(seconds=572, microseconds=750000), value='out'), oneai.Label(type='replacement', skill='enhance', output_spans=[Span(start=41, end=41, section=75, text='')], timestamp=datetime.timedelta(seconds=660, microseconds=770000), value='like'), oneai.Label(type='replacement', skill='enhance', output_spans=[Span(start=12, end=15, section=81, text='I..')], span_text='I..', timestamp=datetime.timedelta(seconds=703, microseconds=730000), value='I.'), oneai.Label(type='replacement', skill='enhance', output_spans=[Span(start=110, end=115, section=83, text='think')], span_text='think', timestamp=datetime.timedelta(seconds=732, microseconds=830000), value='things'), oneai.Label(type='replacement', skill='enhance', output_spans=[Span(start=67, end=71, section=187, text='whom')], span_text='whom', timestamp=datetime.timedelta(seconds=1339, microseconds=130000), value='who'), oneai.Label(type='replacement', skill='enhance', output_spans=[Span(start=15, end=19, section=189, text='whom')], span_text='whom', timestamp=datetime.timedelta(seconds=1349, microseconds=210000), value='who'), oneai.Label(type='replacement', skill='enhance', output_spans=[Span(start=108, end=119, section=193, text='likability.')], span_text='likability.', timestamp=datetime.timedelta(seconds=1383, microseconds=830000), timestamp_end=datetime.timedelta(seconds=1384, microseconds=190000), value='likeability.'), oneai.Label(type='replacement', skill='enhance', output_spans=[Span(start=0, end=3, section=215, text='You')], span_text='You', timestamp=datetime.timedelta(seconds=1516, microseconds=250000), timestamp_end=datetime.timedelta(seconds=1516, microseconds=310000), value='Now you'), oneai.Label(type='replacement', skill='enhance', output_spans=[Span(start=4, end=4, section=241, text='')], timestamp=datetime.timedelta(seconds=1618, microseconds=730000), value='we'), oneai.Label(type='replacement', skill='enhance', output_spans=[Span(start=54, end=54, section=250, text='')], timestamp=datetime.timedelta(seconds=1658, microseconds=30000), value='like')]]

```

Conclusion

From podcasting to Youtube videos to anything that can help your business in the audio and visual marketing world- adding transcripts is important. Adding transcriptions can help boost our SEO and open our business to anyone who might not have the time/ability to watch a video. 

As important as these transcripts are, it’s even more important to make sure that they are read and understood by the audience. Utilizing the proofreading Skill on OneAi’s Language Studio will help both our business practices and our viewers. 

Welcome to the Language Studio, where you can share your thoughts and check out other useful skills.

Perfect Transcripts Effortlessly

Read Next

AI Expert Onboarding Session

Start smart - schedule a free session with one of our AI experts to set up and configure your agent for success