How Speechtext.io adventure started?

As a tech based and innovative company located at Techno Park, we are always motivated to help our users to make their daily base works easier. Our company’s main vision is to develop technologies that really make our users’ life better and easier. In fact, we weren’t intended to develop a service that makes audio or video transcription automated at first.

Figure 1: Techno city Twins Building located at the heart of university campus
Figure 1: Technocity Twins Building located at the heart of university campus

Since we are located at university campus and research centers; especially most social science researchers (including researchers working within the company) are asking us to help them with the audio records such as  survey or questionnaire results, and interview records to be easily converted into text files. Researchers main problem was as follows:

Limitations that Researchers encounter when doing transcription

  • Doing the transcription process manually is time consuming, an hour audio or video record can be transcribed within 6-8 hours.
  • It is hard to find professional transcribers to do transcription for hundreds of hours audio records.
  • Manual transcription has high cost rates and hard to allocate that much money for their research budget.
  • There is a very limited automatic transcription tool or service other than English language. 
  • For Turkish and Arabic language, most transcription services don’t support or the ones that support multiple languages have a very low accuracy rate.
  • Finally for the proofreading process or correcting the transcript, they need a tool that transcripts can be edited.
Transcription can consume a lot of time of researchers
Figure 2: Transcription can consume a lot of time of researchers

How we decided to develop an automatic transcription service?

We have developed automatic transcription service firstly for our researchers especially in Turkish and Arabic language for their quantitative design research. SpeechText.io has been developed specifically for the need to decode audio / video recordings that arise during research projects and graduate studies. Initially, SpeechText.io, which meets the voice decoding needs of the researchers working within the company, has matured in time and has been opened to the service of all users who perform decoding and transcription services manually, social science researchers and media sector employees. Supporting 120 languages and dialects including Turkish, English, Spanish, Arabic and Russian.

Figure 3: Speechtext.io aims to provide productivity tools for converting audio to text
Figure 3: Speechtext.io aims to provide productivity tools for converting audio to text

Who is it for and main features of SpeechText.io?

So the adventure has started within the need of our researchers, continuing to develop the transcription service by adding new features such as online editor (underlying words that have low accuracy rate), subtitle generator etc. We provide transcription service with lowest cost (offered approximately % 40 lower price than alternatives) and also do offer educational discounts to those who provide verification. Other than researchers, we believe that our transcription service will help people who are:

  • Content producers (Youtubers, podcast publisher, video producers etc.)
  • Media sector (Journalists that make interviews )
  • Education (universities that make quantitative design research)
  • Marketers 
  • Freelance Transcribers
  • Subtitlers

Please feel free to contact about any subject by filling our contact form or sending an email to [email protected].

Name (required)Email (required)Website

Leave a Reply