21st
century is a century of revolution in Artificial Intelligence
and Human Machine Interface Systems.
Text To Speech Technology is a branch of Artificial
Intelligence. Text To Speech Synthesis is a voice
technology in which raw text is converted into audible
speech. Text To Speech (TTS) is a process through
which input text is analyzed, processed, and "understood" then
the text is rendered as digital audio and then "spoken".
Most Text To Speech Systems can be categorized by
the method that they use to translate phonemes into
audible sound. Some of them are listed below :
Prerecorded
Formant
Concatenated
We
are proposing a new technique (RecSimCat Technique)
for the generation of Highly Intelligible synthesized
voice with a natural sound.
Prerecorded
Concatenated
Formant
RecSimCat
Resource
requirement
Very
large storage, Small memory
Large
storage, Very large memory
Low
storage, relatively small memory
Low
storage, Low memory
Vocabulary
Limited
Unlimited
Unlimited
Unlimited
Voice
quality
Natural,
Most pleasant
Natural
Robotic,
Sometimes not appreciated by the user
Natural
Multiple
featured voices
Need
high storage in that case
Need
high storage in that case
Can
produce multiple featured voices without any
major changes
Can
produce multiple featured voices without any
major changes
Intelligibility
High
Highly
Intelligible
High
Highly
Intelligible
Features:
Output
Sample Rate and Sample Size of generated speech are
22 khz, 8 bit, mono.
Multiple featured voices of adult female, naughty
child, strong female, adult male etc.
User can vary the speed of pronunciation.
It
can run with this configuration or above:
· Pentium Processor 200 MHz
· OS = Windows 95/98, Me, NT, 2000
· 16 MB RAM
Applications
where Hindi TTS can be used:
Dictation
Systems
Telephony
Unified
Messaging
Information
Kiosks
Reader
Talking
Web Pages
Games/Edutainment
Automobile
Intelligent
Agents
Conclusion:
RecSimCat
Technique has all the combined advantages like need
of low disk space, intelligibility, capability of producing
multiple featured voices, elimination of robotic voice,
ability of reading text from any external window etc.
By the progress of time this technology will gain popularity
because of its simplicity. We can make TTS System for
any language by using RecSimCat Technique.