A DSP processor is a specialized microprocessor with its architecture optimized for the needs of Digital signal processing and our aim for this experiment is to undergo through a Signal processing application.The goal of this group Experiment of fetching out papers and patents on DSPP applications reviewing it and compiling a IEEE formatted report has been achieved and following order flows the report reviewed by Amey C. Thombre and Saumil S. Shah followed by the paper and patent reviewed.
Your Valuable comments and suggestions are much appreciated.
Abstract - this work describes a process of a speech recognition system
as an application of DSP processor designed and developed proportional
and linearly to overcome the holistic technique of voice synthesis and
human to computer interaction with a better, accurate and much
appreciated methodology. The design and implementation of this process
is done using MATLAB at the initial simulation level, a multi-paradigm
numerical computing environment and fourth-generation programming
language software. The project involves deep insights of data processing,
modulation techniques, mathematical analysis and conversion and
generation. The DSP Processor plays the play-making role in this
construction as it is crucially held responsible for the implementation part.
Generally, the input voice is recognized by the system for any type of input.
But here, the apparatus is able to differentiate between various voice
signals as well. This can be used for various security purposes because it
can be trained to recognize a specific set of phrases by the user. Speech
recognition is a very computationally intensive task and includes many of
digital signal processing algorithms to be executed simultaneously which
makes it hard and also beneficial as it increases the speed by parallel
processing. Many Kits like HM2007 use this technique for voice recognition
Harmonics may be present in the sinusoidal output due to non-linear loads.
High harmonic content can cause excessive heating in the circuitry and give
incorrect readings. It is therefore essential that the sinusoidal output
generated has minimum harmonic content.
IEEE Paper Review: DSP Based System for Real time Voice Synthesis Applications Development
Publisher :
Radu Arsinte,Attila Ferencz
-Software ITC S.A.- 109 Gh.Bilascu Street -3400 Cluj-Napoca - Romania Phone:+40-64-197681,197682 Fax:+40-64-196787 Email:sitc@utcluj.ro Email:dianaz@utcluj.ro
-Software ITC S.A.- 109 Gh.Bilascu Street -3400 Cluj-Napoca - Romania Phone:+40-64-197681,197682 Fax:+40-64-196787 Email:sitc@utcluj.ro Email:dianaz@utcluj.ro
Costin Miron Technical University Cluj-Napoca - Faculty of Electronics and Telecommunications 26-28 Gh.Baritiu Street ,3400 Cluj-Napoca,Romania
Summary : This paper describes an experimental system designed for development of real time voice synthesis applications. The system is composed from a DSP coprocessor card , equipped with an TMS320C25 or TMS320C50 chip, voice acquisition module (ADDA2) ,host computer (IBM-PC compatible), software specific tools.
The development system was used in the real time implementation of speech synthesiser based on the linear prediction method. Using DSP technology allows real-time synthesis of voice ,with high quality features. Compared with PC only based systems (without DSP) performances are higher then in a PC486DX4 or Pentium implementation, where is difficult to obtain real-time running with this method(as experiments revealed).On other side in PC throughput limitations occurs, extension bus being considerably slower than the CPU. After the development phase ,done on this system, we can design a dedicated system for consumer applications based on DSP ,the resulting system cost being significantly reduced.
Patent Review :
Adaptation of a speech recognition system across multiple remote sessions with a speaker.Patent No : US 6766295 B1
Filing Date : 20 JUlY 2004
Inventors : Hy Murveit, Ashvin KannanOrignal Assignee : Nuance Commnications
Summary : A technique for adaptation of a speech recognizing system across multiple remote communication sessions with a speaker. The speaker can be a telephone caller. An acoustic model is utilized for recognizing the speaker's speech. Upon initiation of a first remote session with the speaker, the acoustic model is speaker-independent. During the first session, the speaker is uniquely identified and speech samples are obtained from the speaker. In the preferred embodiment, the samples are obtained without requiring the speaker to engage in a training session. The acoustic model is then modified based upon the samples thereby forming a modified model. The model can be modified during the session or after the session is terminated. Upon termination of the session, the modified model is then stored in association with an identification of the speaker. During a subsequent remote session, the speaker is identified and, then, the modified acoustic model is utilized to recognize the speaker's speech. Additional speech samples are obtained during the subsequent session and, then, utilized to further modify the acoustic model. In this manner, an acoustic model utilized for recognizing the speech of a particular speaker is cumulatively modified according to speech samples obtained during multiple sessions with the speaker. As a result, the accuracy of the speech recognizing system improves for the speaker even when the speaker only engages in relatively short remote sessions.
The development system was used in the real time implementation of speech synthesiser based on the linear prediction method. Using DSP technology allows real-time synthesis of voice ,with high quality features. Compared with PC only based systems (without DSP) performances are higher then in a PC486DX4 or Pentium implementation, where is difficult to obtain real-time running with this method(as experiments revealed).On other side in PC throughput limitations occurs, extension bus being considerably slower than the CPU. After the development phase ,done on this system, we can design a dedicated system for consumer applications based on DSP ,the resulting system cost being significantly reduced.
Patent Review :
Adaptation of a speech recognition system across multiple remote sessions with a speaker.Patent No : US 6766295 B1
Filing Date : 20 JUlY 2004
Inventors : Hy Murveit, Ashvin KannanOrignal Assignee : Nuance Commnications
Summary : A technique for adaptation of a speech recognizing system across multiple remote communication sessions with a speaker. The speaker can be a telephone caller. An acoustic model is utilized for recognizing the speaker's speech. Upon initiation of a first remote session with the speaker, the acoustic model is speaker-independent. During the first session, the speaker is uniquely identified and speech samples are obtained from the speaker. In the preferred embodiment, the samples are obtained without requiring the speaker to engage in a training session. The acoustic model is then modified based upon the samples thereby forming a modified model. The model can be modified during the session or after the session is terminated. Upon termination of the session, the modified model is then stored in association with an identification of the speaker. During a subsequent remote session, the speaker is identified and, then, the modified acoustic model is utilized to recognize the speaker's speech. Additional speech samples are obtained during the subsequent session and, then, utilized to further modify the acoustic model. In this manner, an acoustic model utilized for recognizing the speech of a particular speaker is cumulatively modified according to speech samples obtained during multiple sessions with the speaker. As a result, the accuracy of the speech recognizing system improves for the speaker even when the speaker only engages in relatively short remote sessions.
Patent Reviewed : https://drive.google.com/open?id=0BxKOmgoubcmEdlZxNVhySTZGSGc
Paper Reviewed : https://drive.google.com/open?id=0BxKOmgoubcmEdnAzUEJQQ1BXeHc
Plagarism Report : https://drive.google.com/open?id=0BxKOmgoubcmEdlEySGxzSGJOZWM
