Speech to Text & Text to Speech
Today, content production in business has a high value and has various applications for writers, journalists, lawyers, doctors, public relations departments of organizations and so on. You will be able to convert all your audio files to text with the help of Farsava software and its web panel This software has special features such as editing, saving files, sharing, the ability to receive audio files in various formats, convert it to text and also measure the accuracy of converting files.
Programmers and developers can also add Farsava service, i.e. the ability to convert speech to text and text to speech to their websites and applications.
One of Farsava services is text-to-speech that can receive given texts and convert them to speech. In the latest version of this service, the type of speech broadcast is very close to the speech of a real human being. One of the most important uses of this service is to use it in routing applications to read addresses and routing instructions.
In many cases, instead of recognizing all the words expressed in the audio files, it is necessary to recognize certain words. For this purpose, a product has been produced for keyword extraction (Keyword spotting). This product is able to recognize and extract keywords expressed in audio files with more accuracy (20%) than converting speech to text.
If organizational customers need to install and use Farsava product on the internal server, the basis for pricing is the amount of processing and hardware required to convert speech to text, which will be announced to the customer after the initial expertise.
You will only pay once by purchasing the Farsava package (as a software package on the customer’s internal server) and you will not need to pay for the service on a monthly basis.
Farsava is a software for converting speech to text that provides the possibility of human verbal communication with a computer or mobile phone This software has been produced based on the knowledge of Iranian artificial intelligence specialists and using the latest technologies in the world. it has succeeded in competing with large companies such as Google with a high accuracy factor due to the collection of the largest audio database available in Persian, while benefiting from a very wide variety of speech. Farsava also has two other products called Text to Speech Conversion (TTS) and Keyword Spotting (KWS).
This service is able to convert incoming texts to audio In the latest version of TTS, the type of speech broadcast is very close to that of a real human.
Keyword spotting can understand audio files more accurately (up to 20%) than converting speech to text and recognize keywords in a file and extract them. In addition, this product is able to search for a number of keywords given to it in a collection of multimedia files and also detect when each word occurs in the files.
Farsava product can be supplied in two ways according to customer needs:
1- Providing API / SDK based on internal cloud computing
2- Providing software package in the customer’s internal server
You can send an email to support@amerandish.com to contact the technical support unit
You can refer to the following address and discuss your problems in the issue section.
https://github.com/amerandish
Newsletter subscription
Newsletter subscription