Anryze is a project that plans on using speech recognition platform with distributed computing software. It is a project that aims to change the current speech recognition technology.
Attempts have been made over the past 50 years by several companies, including Bell Labs and IBM, to create a program that would allow computers to understand the human language. This technology would allow us to tap into a higher number of features and increase the responsivity of computers in our day to day life.
So far the most advanced speech recognition technology has been introduced by Google, in the form of their Speech API. However, Anryze plans to take it to a more advanced level in order to keep up with the needs of the tech-savvy generation.
How it works
In order to make something this complex work, Anryze has employed an advanced mechanism.
The first step includes recognizing the sounds. This is done by taking the waveform, splitting it on utterances by silences then trying to recognize what is being said in each utterance. This is followed by taking all possible combinations of words and trying to match them with the audio. The best matching combination is then selected using various models.
For the process of matching, three different models are used. This includes the acoustic model which contains the acoustic property for each sound, the phonetic model which contains the mapping from words to phones and the language model which is used to restrict word search. These are then combined together in order to create a program that would enable speech recognition.
The audio file that needs to get converted is uploaded by the user and the transaction cost is calculated depending upon the size and the length of the file.
Benefits
Through their advanced technology, Anryze wants to create a speech recognition system that is both responsive and accurate
The main problem that they aim to tackle is precision. They have employed optimized software that will enable them to implement accurate and efficient searches. They want to allow the computers to achieve more than they currently do through speech recognition.
In order to support their study, the team of Anryze carried out tests to compare the mistakes made in the speech recognition systems of MS Azure, Google Speech, IBM Watson and Anryze STT V.b0.53. The results showed that the percentage of mistakes while recognizing the speech was comparatively lower for Anryze.
Token and ICO
Anryze will use their own unique token called RYZ to carry out transactions on their platform. Users will be able to pay for services through these tokens and miners will be paid in return for their services in RYZ as well. Investors will also be paid their commission through these tokens.
Their aim is to raise $1.6–6 million by selling tokens during their pre-ICO and ICO. However, if they fail to raise this amount, money would safely be returned back to contributors and the company will still continue to work on their ambitious project.
Review and Conclusion
According to Anryze, the global voice recognition market size was valued at $51.09 billion in 2015 and is expected to grow at 11% per year. There is no denying that the ever-growing need for development in technology will ensure a bright future for the project.
The barrier of a computer not understanding human language is definitely the main drawback that exits in the world. With the right technology in place, we would be able to promote further advancement in the education sector, in the healthcare sector, for military purposes, and would even in the technology of smartphones. The possibilities for growth would be endless.
This project shows immense potential for growth and definitely has a promising future ahead.
For more information visit:
Website: https://tokensale.anryze.com/
Written By: Kishan
Bitcointalk Name: Kishan748
