Speech recognition, also known as automatic speech recognition, allows a computer to understand human voice and perform tasks. It uses acoustic and language models to recognize speech. Acoustic models are statistical representations of sounds created from audio recordings and transcriptions, while language models predict word sequences. There are two main types: speaker-dependent systems require user training to recognize individual voices more accurately, while speaker-independent systems used in applications like phones do not require training but are generally less accurate. The speech recognition process involves digitizing speech, analyzing acoustic signals, and linguistically interpreting the speech to recognize words.

2.  What is Speech Recognition?
Also known as automatic speech
recognition or computer speech
recognition which means
understanding voice by the computer
and performing any required task.

3.  Where can it be used?
 System control/navigation
e.g. GPS-connected digital maps: “How far is it
to the motorway junction?”
 Commercial/Industrial applications
in-car steering systems
 Voice dialing
hands-free use of mobile in car e.g. “Dial
office”

4. Voice Input Analog to Digital Acoustic Model
Language Model
Feedback Display Speech Engine

5.  Acoustic Model
 An acoustic model is created by taking audio recordings of
speech, and their text transcriptions, and using software to
create statistical representations of the sounds that make up
each word. It is used by a speech recognition engine to
recognize speech.
 Language Model
 Language model is used in many natural language
processing applications such as speech recognition tries to
capture the properties of a language, and to predict the next
word in a speech sequence.

6.  Two types of speech recognition.
 Speaker-Dependent
is commonly used for dictation software
 Speaker-Independent
is more commonly found in telephone
applications.

7.  Speaker-dependent software works by learning
the unique characteristics of a single person’s
voice, in a way similar to voice recognition.
 New users must first “train” the software by
speaking to it, so the computer can analyze
how the person talks.
 This often means users have to read a few
pages of text to the computer before they can
use the speech recognition software.

8.  Speaker-independent software is designed to recognize
anyone’s voice, so no training is involved.
 This means it is the only real option for applications
such as interactive voice response systems — where
businesses can’t ask callers to read pages of text before
using the system.
 The downside is that speaker-independent software is
generally less accurate than speaker-dependent
software.
 Speech recognition engines that are speaker
independent generally deal with this fact by limiting
the grammars they use. By using a smaller list of
recognized words, the speech engine is more likely to
correctly recognize what a speaker said.

9.  Articulation produces
 sound waves which
 the ear conveys to the brain
 for processing

10. Acoustic waveform Acoustic signal
 Digitization
 Acoustic analysis of the
Speech recognition
speech signal
 Linguistic interpretation

12. • Digitization
 Analogue to digital conversion
 Sampling and quantizing
 Sampling is converting a continuous signal into a discrete signal
 Quantizing is the process of approximating a continuous range of
values
• Signal processing
– Separating speech from background noise
• Phonetics
– Variability in human speech
• Phonology
– Recognizing individual sound distinctions (similar
phonemes)
– is the systematic use of sound to encode meaning
in any spoken human language

13.  Semantics and pragmatics
 Semantics tells about the meaning
 Pragmatics is concerned with bridging the
explanatory gap between sentence meaning and
speaker's meaning
 Lexicology and syntax
 Lexicology is that part of linguistics which
studies words, their nature, and meaning.
 Syntax tell about the arrangement of words and
phrases to create well-formed sentences.

14. S1
Speaker Speech
parsing S2
and
Recognition Recognition
arbitration
SK
SN

15. Switch on S1
Channel 9
Speaker Speech
parsing S2
and
Recognition Recognition
arbitration
SK
SN

16. Who is S1
speaking?
Speaker Speech
parsing S2
and
Recognition Recognition
arbitration
SK
Annie
David SN
Cathy
“Authentication”

17. What is he S1
saying?
Speaker Speech
parsing S2
and
Recognition Recognition
arbitration
SK
On,Off,TV SN
Fridge,Door
“Understanding”

18. What is he
talking S1
about?
Speaker Speech
parsing S2
and
Recognition Recognition
arbitration
SK
“Switch”,”to”,”channel”,”nine”
Channel->TV
SN
Dim->Lamp
On->TV,Lamp
“Inferring and execution”

19.  http://www.slideshare.net/richiebmthimmaia
h/automatic-speech-recognition-4721204
 http://www.scribd.com/doc/36605017/Voice
-Recognition-System-PPT
 http://www.google.com.pk/url?
sa=t&rct=j&q=speech%20recognization
%20system%20.ppt
 http://www.wikipedia.com