A comparative study of pitch detection algorithms for microcontroller based voice pitch detector
This paper presents a study to compare the performance of two pitch detection algorithms namely the Autocorrelation Function and the Cepstrum Analysis to select a suitable algorithm that can be developed into a standalone voice pitch detector. The two algorithms were chosen due to their uncomplicate...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | en |
| Published: |
American Scientific Publishers
2017
|
| Subjects: | |
| Online Access: | https://eprints.ums.edu.my/id/eprint/45810/1/FULLTEXT.pdf https://eprints.ums.edu.my/id/eprint/45810/ https://doi.org/10.1166/asl.2017.10320 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | This paper presents a study to compare the performance of two pitch detection algorithms namely the Autocorrelation Function and the Cepstrum Analysis to select a suitable algorithm that can be developed into a standalone voice pitch detector. The two algorithms were chosen due to their uncomplicatedness to be realized in a microcontroller. The performance of both algorithms was analyzed using 288 speech samples recorded from 24 students in both quiet and noisy environments. Results showed that both algorithms produced comparable pitch values and were able to determine the pitch of speech signals correctly. However, in terms of complexity and computational processing time, the Autocorrelation Function performed better than the Cepstrum Analysis. |
|---|