The Battle of Speech Technology: Google VS MySteno
We all have had to deal with someone or the other at our workplace or college who talks too fast for us to follow. Having to transcribe the notes of someone like that is a real pain, and that’s why text-to-speech technology has made all our lives easier. It helps us save valuable time and effort by helping us collect the required information faster than ever before. Tech giants Google and Amazon are spearheading this area of science, but a lot of new entries are now coming into this massively useful field which will determine the future of voice- recognition systems. In this edition, we compare two products by two different companies, one being Google and the other being Vakil Search who have released their own transcribing tool- MySteno.
The Companies
- Google is a tech giant that needs no introduction, and while it has been around for a while, its interest in voice technology started nearly a decade ago. The Google Voice assistant has become a worldwide success and help several millions of people around the world by making it easier to play a song, or write a note. Google has had to face stiff competition from Amazon, Microsoft and even Apple as they all entered the filed with their consequent voice-to-text applications such as Siri, Cortana and Alexa. Hence, the market share has now been divided between Google and Amazon as they have proven to be the most successful in this field. The issue that such companies face is that this is just a subsidiary product they offer and is hence not their primary concern. Every year, they have to find funds to ump into the R&D required for such applications from other sources, and they also have to split their resources and staff to field multiple projects in multiple areas of technology.
- Vakil Search, meanwhile is a totally different organisation that was set up by a lawyer with an intention to help other lawyers and provide them with services that will help significantly in reducing their workload and simplifying their lives. The company has already proven its tech-savvy nature by coming out with Libra- The Advocates App, which was a resounding success due to its ability to organise the work life of lawyers around the world. It second venture into the tech field is MySteno, which is a voice-to-text transcription application that may be accessed through their website.
Languages
While Google Speech offers a wide range of languages, providing support for over 119 languages which include thirteen variations of the English such as Australian, Canadian and Indian, VS’s MySteno has instead decided to focus on just English.
Google also supports nine Indian languages, namely:
- Bengali
- Hindi
- Gujarati
- Kannada
- Malayalam
- Marathi
- Tamil
- Telugu
- Urdu
While supporting more languages increases scope and visibility, it also adds strain on the developers as they have to maintain the same quality in translation and transcription for all these different languages. This isn’t always easy as certain algorithms work better with individual specific languages.
Audio Size
Google Speech employs separate systems for different lengths and sizes of audio it receives. Therefore, it prefers a proper transcription style for a long speech, whereas the short speech style works best for voice interfaces. This is one of the significant differences that it has with other tech giants as even Amazon uses a common input channel for all types of audios coming into the transcriber. But both Amazon and Google’ have an established input audio limit that stands at 120 minutes per API call. MySteno employs the same mechanism for most of its note-taking functionalities.
Privacy
This is one section in which both match up as both Google Speech and MySteno value the customer’s privacy. Both allow for data privacy by preventing the leakage of data. Google further uses data received from customers to adapt and evolve its software and even uses it to improve its machine learning models.
Speed
Both are super efficient and super fast. The Mysteno app takes just 20 minutes to wind up a ten-page document which would take manual transcribers around 2 hours to take down. It types as you talk, helping you save so much time and effort, both of which can be used for more important things.
As a whole, both the platforms are equally competitive. While Google may beat MySteno in terms of the number of supported languages, it does not feature a custom vocabulary and punctuation option like MySteno, making them different and unique at the same time. In certain respects, they are quite alike, such as with respect to the basic functionality and formats. As the MySteno app is built keeping lawyers in mind, it has several additional features such as the option to dictate punctuations using AI whenever required, whereas Google does not support this functionality. It’s AI also stores over 2000 Legal abbreviations and maxims making it easier than ever before to draft legal documents. It also has the extra feature of providing the user with the chance to convert the word document to a PDF which can be readily presented, making it the best app for advocates and law students.
More or less, both platforms offer multiple advantages to manual transcription and are hence helping in advancing voice-to-text technology. It will be interesting to see how they match up in the future and to see whether new tech will help one overpower the other.