A Speech Recognition (SR) is a technique to identify audible words and phrases in spoken languages and convert them to machine-readable format.
The whole concept of Speech Recognition works mainly bases on two things one that is how accurately the program reads the audible voices and second is how much the database is accurate to handle a multiple languages and multiple voices, These two are major components to develop a perfect Speech Recognition System.
What can we do with Speech Recognition?
Libraries for Speech Recognition in iOS
There are several commercial libraries as well as Open Source libraries are available to implement SR in iOS applications. All libraries have it’s own advantages and disadvantages with respect to it’s functionality, performance, cost, stability, capability etc. Following are list of some libraries that are available for iOS.
1. PocketSphinx:
Pocket Sphinx is the Open Source library. This library has lots of advantages as compare other libraries, It provides support for both desktop apps as well as mobile apps, Provides a cross-platform supports like Linux, Windows, Mac OS X, iOS, python language binding, even experimental support for Nokia S60v3 and Windows Mobiles.
Implementation of PocketSphinx is quite difficult as compare to other libraries as it has not proper documentation support for developers.
2. VocalKit:
VocalKit is a free library and wrapper for already available library (Pocket Sphinx), main aim to develop this library is to provide ease to use voice recognition solutions, but now a days VocalKit is not preferable to use as much better library openEars came out.
3. OpenEars:
openEars is a free library for offline Speech Recognition and Text To Speech. The library has no any hidden costs or account to setup. OpenEars makes it simple to add Speech Recognition by providing a well-written document, openEars featuring high performance, low CPU usage, multiple voices and more.
openEars also provides some paid plugins support for several new speech features into your application to provides a live speech recognition in realtime. It is very much responsive that it can be used as a gaming input without the use of network connection, NeatSpeech is a another plugin for openEars which add improved voices which sound better and continuously even on long sentences, Rejecto is a vocabulary rejection plugin for the words and noises which aren’t in the app’s vocabulary, SaveThatWave is a plugin which adds a capability to record audio files from speech.
4. iVona:
iVona is a paid library provides a reach documentation help and tutorials about how to use it’s library. It provides a SDK for variety of platforms like Linux, iOS, Windows mobile, Mac OS, and etc. iVona provides a support for variety of languages like Tatyana, Mads etc.
iOS 7 & Speech Recognition:
Apple has provides a “1500 new APIs” with NDA of iOS7, One of them is AVSpeechSynthesier. AVSpeechSynthesier is Text To Speech that is a part of AVFoundation framework and provides a very easy way to have iOS read a piece of text.
AVSpeechSynthesizer *synthesizer = [[AVSpeechSynthesizer alloc] init]; AVSpeechUtterance *utterance = [AVSpeechUtterance speechUtteranceWithString:@"Hey, how are you?"]; utterance.rate = AVSpeechUtteranceMinimumSpeechRate; // Tell it to me slowly [synthesizer speakUtterance:utterance];
Enjoy to make an iPhone an iPad application that fun with voice. SPEC INDIA is a leading mobile apps development company offering different custom applications, enterprise mobile app development for global clients. Our skilled mobility expert team delivers high quality, effective, efficient and user-friendly applications that meet our client’s expectations.
The whole concept of Speech Recognition works mainly bases on two things one that is how accurately the program reads the audible voices and second is how much the database is accurate to handle a multiple languages and multiple voices, These two are major components to develop a perfect Speech Recognition System.
 |
| Speech Recognition in iOS Application |
- Speech Recognition provides a different level of experience to user, User can have a full control of system by speaking just a several voice commands and the entire system will work without a press of any single button or without a touch of single thing.
- Opens an accessibility path for the deaf & hard hearing.
- Person is free to use SR System at any time while driving, cooking, jogging etc.
- SR can reach a gaming world to some another direction instead of pressing a single key, user can have just a simple voice commands to interact with game this provides a real-time SR, in Gaming world.
- Speech Recognition can be useful for Identification of a particular user by implementing intelligent SR System, Which scans user’s voice and provide an identification of user.
- The application created based on SR can be used as Virtual Assistance like a Siri application in iOS.
Libraries for Speech Recognition in iOS
There are several commercial libraries as well as Open Source libraries are available to implement SR in iOS applications. All libraries have it’s own advantages and disadvantages with respect to it’s functionality, performance, cost, stability, capability etc. Following are list of some libraries that are available for iOS.
1. PocketSphinx:
Pocket Sphinx is the Open Source library. This library has lots of advantages as compare other libraries, It provides support for both desktop apps as well as mobile apps, Provides a cross-platform supports like Linux, Windows, Mac OS X, iOS, python language binding, even experimental support for Nokia S60v3 and Windows Mobiles.
Implementation of PocketSphinx is quite difficult as compare to other libraries as it has not proper documentation support for developers.
2. VocalKit:
VocalKit is a free library and wrapper for already available library (Pocket Sphinx), main aim to develop this library is to provide ease to use voice recognition solutions, but now a days VocalKit is not preferable to use as much better library openEars came out.
3. OpenEars:
openEars is a free library for offline Speech Recognition and Text To Speech. The library has no any hidden costs or account to setup. OpenEars makes it simple to add Speech Recognition by providing a well-written document, openEars featuring high performance, low CPU usage, multiple voices and more.
openEars also provides some paid plugins support for several new speech features into your application to provides a live speech recognition in realtime. It is very much responsive that it can be used as a gaming input without the use of network connection, NeatSpeech is a another plugin for openEars which add improved voices which sound better and continuously even on long sentences, Rejecto is a vocabulary rejection plugin for the words and noises which aren’t in the app’s vocabulary, SaveThatWave is a plugin which adds a capability to record audio files from speech.
4. iVona:
iVona is a paid library provides a reach documentation help and tutorials about how to use it’s library. It provides a SDK for variety of platforms like Linux, iOS, Windows mobile, Mac OS, and etc. iVona provides a support for variety of languages like Tatyana, Mads etc.
iOS 7 & Speech Recognition:
Apple has provides a “1500 new APIs” with NDA of iOS7, One of them is AVSpeechSynthesier. AVSpeechSynthesier is Text To Speech that is a part of AVFoundation framework and provides a very easy way to have iOS read a piece of text.
AVSpeechSynthesizer *synthesizer = [[AVSpeechSynthesizer alloc] init]; AVSpeechUtterance *utterance = [AVSpeechUtterance speechUtteranceWithString:@"Hey, how are you?"]; utterance.rate = AVSpeechUtteranceMinimumSpeechRate; // Tell it to me slowly [synthesizer speakUtterance:utterance];
Enjoy to make an iPhone an iPad application that fun with voice. SPEC INDIA is a leading mobile apps development company offering different custom applications, enterprise mobile app development for global clients. Our skilled mobility expert team delivers high quality, effective, efficient and user-friendly applications that meet our client’s expectations.
Posts
0 comments:
Post a Comment