Skip to content
This repository has been archived by the owner on Sep 10, 2022. It is now read-only.

Latest commit

 

History

History
64 lines (62 loc) · 3.82 KB

ReadMe.md

File metadata and controls

64 lines (62 loc) · 3.82 KB
Raw

Speech Processing CS 566

Assignment 03 (Vowel Recognition)

Roll No: 214101058 MTech CSE'23 IITG | Vijay Purohit

Input: 
	*	Vowel Data Files '.txt' in format 214XXXXXX_%c_%d.txt (c={a,e,i,o,u} d={1,2,3 ..., 20} 100 files)
	*	Kept inside the path provided by array filePathVowels[] = "input_vowels_data/". (Line #89)
	*	Roll Num can be changed by variable rollNum[]="214XXXXXX" (Line #55)
Output:  
	*	R1:Final Reference Files for each vowel by analysing the first 10 files of each vowel.
			Folder: fileOutputRef[] = "output_reference_file/";
			Format: {214XXXXXX} + _ + {a} + _avg_ci_reference.txt
	*	R2:Collective Analysed Files Data for each vowel by analysing the first 10 files in single file for each vowel.
			Folder:  fileOutputRef[] = "output_reference_file/";	
			Format: {214XXXXXX}+ _ + {a} + _final_analysis.txt
	*	T1:Individual Test File Result after comparing with each reference file of vowel.
			Folder: fileOutputTest[] = "output_test_result/";
			Format: {214XXXXXX} + _ + {a} + {1} + _test_result.txt
	*	T2:Collective Test File Result of all the test files (last 10 files of each vowel). 
			Folder: fileOutputTest[] = "output_test_result/";
			Format:  _ + {214XXXXXX} +_aeiou_TEST_RESULT_OUTPUT.txt
	*   (optional when var segregate_speech=TRUE (Line #93))
			Segregated Speech data of each file analysed for testing/debugging purpose. Can be played in Cool Edit.
			Folder:  fileOutputRecording[] = "output_recordings/";
			Format: {214XXXXXX} + _ + {a} + _ + {1} + _normalized_segregated_recording.txt
Details About Each File:
	*   	R1:Format: {214XXXXXX} + _ + {a} + _avg_ci_reference.txt
			Contains final Avg Ci (5*12) of first 10 files of each vowel.
			Total Five Files.
	*   	R2:Format: {214XXXXXX}+ _ + {a} + _final_analysis.txt
			It contains values of DC shift, normalization factor, STE marker Frame of each 10 files.
			Contains Coefficients Values of each 10 files and For each 5 frame.
			Finally the Avg C_i values computed from 10 files in proper readable format.
			Total Five Files.
	*   	T1:Format: {214XXXXXX} + _ + a + {1} + _test_result.txt
			It contains values of DC shift, normalization factor, STE marker Frame of test file.
			Contains Coefficients Values of each 5 frame.
			Finally the tokhura distance calculated with respect to each frame and each vowel reference file.
			Result of minimum tokhura distance and vowel recognized.
			One for each test file.
	*   	T2:Format:  _ + {214XXXXXX} +_aeiou_TEST_RESULT_OUTPUT.txt
			Console Output of all the test file result. (10files*5vowels)
			Result = Their Tokhura Distance and vowel recognized. File Testing data path.
			Accuracy with respect to 10 vowel files and overall accuracy in the end.
			Single file.

Instructions to execute Code.

  1. Open it in Visual Studio 2010. Main file: SP_AS03.cpp
  2. Input format is given above.
    • Change Roll Number Variable rollNum[](Line #55), if required for different files.
    • Remember vowel data should be present in folder provided by array filePathVowels[] (Line #89).
    • Make Sure files are in proper format as mentioned. Otherwise some logic at (Line #679) needed to be modified.
  3. Compile it and Run. Console window will show up.
    • First is the Validation of Calculation of Coefficients. Press Enter.
    • Second is the Generation of Reference file for Each vowel. Press Enter to generate for Each Vowel. (Folder by: fileOutputRef[])
    • Third is the Testing of Remaining file of each vowel. Press Enter to test for Each Vowel. (Folder by: fileOutputTest[])
    • Final Output will be shown in console.
    • As well as it is saved in file [_ + {214XXXXXX} +_aeiou_TEST_RESULT_OUTPUT.txt] in (Folder by: fileOutputTest[])

THE END.