Ai in speech recognition

Artificial Intelligence and voice recognition

Artificial Intelligence and voice recognition

Himanshu ChoubisaHimanshu Choubisa

Lets Define !Lets Define !

“Artificial intelligence is the science and engineering of making intelligent machines, especially intelligent computer programs. It is related to the similar task of using computers to understand human intelligence, but AI does not have to confine itself to methods that are biologically observable.

“Artificial intelligence is the science and engineering of making intelligent machines, especially intelligent computer programs. It is related to the similar task of using computers to understand human intelligence, but AI does not have to confine itself to methods that are biologically observable.

Speech RecognitionSpeech Recognition

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.

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.

How does it work?How does it work?


• Signal processing-Convert the audio wave into a sequence of

feature vectors• Speech recognition:-Decode the sequence of feature vectors into a

sequence of words

• Signal processing-Convert the audio wave into a sequence of

feature vectors• Speech recognition:-Decode the sequence of feature vectors into a

sequence of words


• Semantic interpretation:

-Determine the meaning of the recognized words

• Dialog Management:-Correct errors and help get the task done

• Semantic interpretation:

-Determine the meaning of the recognized words

• Dialog Management:-Correct errors and help get the task done


• Response Generation

-What words to use to maximize user understanding

• Speech synthesis (Text to Speech):

Generate synthetic speech from a ‘marked-up’ word string

• Response Generation

-What words to use to maximize user understanding

• Speech synthesis (Text to Speech):

Generate synthetic speech from a ‘marked-up’ word string

Type of speech recognitionType of speech recognition

-Speech recognition systems can be separated in several different classes by describing what types of utterances they have the ability to recognize. These classes are classified as the following:

-Speech recognition systems can be separated in several different classes by describing what types of utterances they have the ability to recognize. These classes are classified as the following:


Isolated Words:

Isolated word recognizers usually require each utterance to have quiet (lack of an audio signal) on both sides of the sample window.

Isolated Words:

Isolated word recognizers usually require each utterance to have quiet (lack of an audio signal) on both sides of the sample window.


• Connected Words:

-Connected word systems (or more correctly 'connected utterances') are similar to isolated words

• Connected Words:

-Connected word systems (or more correctly 'connected utterances') are similar to isolated words


Continuous Speech:

-Continuous speech recognizers allow users to speak almost naturally

Continuous Speech:

-Continuous speech recognizers allow users to speak almost naturally


• Spontaneous Speech:

-At a basic level, it can be thought of as speech that is natural sounding and not rehearsed

• Spontaneous Speech:

-At a basic level, it can be thought of as speech that is natural sounding and not rehearsed

Advantages of speech recognition


• Work processes become more efficient because document processing times become shorter. Documents can be generated up to three times as fast with speech recognition as they can if they are typed

• Work processes become more efficient because document processing times become shorter. Documents can be generated up to three times as fast with speech recognition as they can if they are typed



• The employment of speech recognition software saves a great deal of labour, particularly for the secretarial staff, which as a rule only needs to make minor corrections to documents

• The employment of speech recognition software saves a great deal of labour, particularly for the secretarial staff, which as a rule only needs to make minor corrections to documents



• The speech recognition software learns as it works if its recognition errors are corrected. This allows the recognition rate to be improved even further.

• The speech recognition software learns as it works if its recognition errors are corrected. This allows the recognition rate to be improved even further.



• Speech recognition software allows dictations from digital dictation devices to be effortlessly transformed into text. For further information, read our article entitled “Dictation and Speech Recognition”.

• Speech recognition software allows dictations from digital dictation devices to be effortlessly transformed into text. For further information, read our article entitled “Dictation and Speech Recognition”.



• Free up cognitive working space;

• Allows dictation of text, commands;

• Eliminates handwriting, spelling problems;

• Always spells correctly (doesn't always recognize words correctly);

• Free up cognitive working space;

• Allows dictation of text, commands;

• Eliminates handwriting, spelling problems;

• Always spells correctly (doesn't always recognize words correctly);



• Those who engage in mobile dictation need scarcely change their routine to take advantage of speech recognition.

• Allows user to operate a computer by speaking to it;

• Those who engage in mobile dictation need scarcely change their routine to take advantage of speech recognition.

• Allows user to operate a computer by speaking to it;

DisadvantagesDisadvantages

• Even the best speech recognition systems sometimes make errors. If there is noise or some other sound in the room (e.g. the television or a kettle boiling), the number of errors will increase.

• Even the best speech recognition systems sometimes make errors. If there is noise or some other sound in the room (e.g. the television or a kettle boiling), the number of errors will increase.


• Speech Recognition works best if the microphone is close to the user (e.g. in a phone, or if the user is wearing a microphone). More distant microphones (e.g. on a table or wall) will tend to increase the number of errors

• Speech Recognition works best if the microphone is close to the user (e.g. in a phone, or if the user is wearing a microphone). More distant microphones (e.g. on a table or wall) will tend to increase the number of errors


• Requires large amounts of memory to store voice files;

• Difficult to use in classroom settings, due to noise interference;

• Requires each user to train software to recognize voice, hard for poor decoders;

• Requires large amounts of memory to store voice files;

• Difficult to use in classroom settings, due to noise interference;

• Requires each user to train software to recognize voice, hard for poor decoders;


• Makes errors, can be frustrating without adequate support;

• Assists with one stage of the writing process, not a solution to the writing problem.

• Makes errors, can be frustrating without adequate support;

• Assists with one stage of the writing process, not a solution to the writing problem.

Voice RecognitionVoice Recognition• Speech input

– Frequency– Duration – Cadence

• Neutral tone • User friendly

• Speech input – Frequency– Duration – Cadence

• Neutral tone • User friendly

Thank You!!Thank You!!

Ai in speech recognition

Documents

synthetic speech

continuous speech recognizers

recognition errors

recognition rate

sequence of words

dictation of text

mobile dictation

signal processing