You are on page 1of 13

SPEECH RECOGNITION

BY SHUBHAM NAUTIYAL
Contents
 Introduction
 What is speech recognition?
 How it works
 Components
 Benefits
 Drawbacks
 Remaining Problems
 Conclusion
 References
Introduction

 Speech recognition is the process of converting an acoustic signal,


captured by a microphone or a telephone, to a set of words.
 The recognized words can be an end in themselves, as for applications
such as commands & control, data entry, and document preparation.
 They can also serve as the input to further linguistic processing in order
to achieve speech understanding.
What is speech recognition?

 Speech Recognition (SR) is the ability to translate a dictation or


spoken word to text.
 Also known as “automatic speech recognition” (ASR), “computer
speech recognition”, or “speech to text” (STT)
Components

 Audio input
 Grammar
 Speech Recognition Engine
 Acoustic Model
 Recognized text
ASR Model
How it works
Applications

 Online Subtitle creation


 A. I. assistants like google assistant, siri, Amazon Alexa
 Hearing Aids
 Voice dialing
Benefits

 Security
 Productivity
 Advantage for handicapped and blind
 Usability of other languages increases
 Personal voice macros can be created
Drawbacks

 If the system has to work under noisy environments, background noise


may corrupt the original data and leads to SS misinterpretation.
 If words that are pronounced similar for example, their, there, this
technology face difficulty in distinguishing them.
Remaining Problems

 Robustness – graceful degradation, not catastrophic failure


 Portability – independence of computing platform
 Confidence Measures – better methods to evaluate the absolute
correctness of hypotheses.
 Out-of-Vocabulary (OOV) Words – Systems must have some method of
detecting OOV words, and dealing with them in a sensible way.
 Accent, dialect and mixed language – non-native speech is a huge
problem, especially where code-switching is commonplace
References

 www.google.com
 www.wikipedia.com
Thank you

Any questions

You might also like