I am looking at working broadly in the area of Speech Signal Processing, and would be grateful if someone can clarify these doubts!

1) Are there any differences between Processing and Analysis of Speech Signals, in terms of the scope they provide for research? Arent both of them concerned in detecting the spoken language from signal?

2) What are the tools that one will be using in these areas?



closed as too broad by jonsca, nispio, Peter K. Nov 14 '13 at 22:05

Please edit the question to limit it to a specific problem with enough detail to identify an adequate answer. Avoid asking multiple distinct questions at once. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.

  • $\begingroup$ This is just too broad a question to be usefully answered here. Try asking something more specific. $\endgroup$ – Peter K. Nov 14 '13 at 22:07

1) Questions of terminology like that are always debatable! I would say that the term "analysis" describes techniques to convert the speech signal into another representation, closer to the semantic level than a raw list of samples (such as a written transcription in the case of speech recognition, a visualization of the tone curve in the case of a system assisting Mandarin learner, the identity of the speaker in case of speaker identification system, a category of emotion in case of an emotion detection system) ; while "processing" describes problems in which both the input and the output are speech (speech denoising/enhancement, coding and decoding, source separation, 'creative' effects like pitch/timing modifications). Of course, there are contexts in which "processing" describes the whole field, analysis included.

2) Deterministic signal processing (filtering, transforms), statistical signal processing (auto-regressive models, sinusoidal model estimation), statistical modeling and machine learning, psychoacoustics, knowledge of how humans produce speech... If you venture into recognition: language theory (finite state machines), natural language processing, some elements of distributed computing (for large model training), data structures/algorithms. If you venture into speech coding or enhancement applications: information theory, experiment design (for subjective evaluation campaign), DSPs/embedded systems with hardware acceleration or even in some cases hardware design. Tools and language: mostly matlab or scipy for research, mostly C/C++ for implementation; and a scripting language (perl, python) is always helpful when you have to deal with texts/annotations.


Not the answer you're looking for? Browse other questions tagged or ask your own question.