| Autor: |
Azlina Bujang, Ketty Chachil, Wan Ana Zuriana Wan Ismail, Siti Salwa Salleh |
| Rok vydání: |
2018 |
| Předmět: |
|
| Zdroj: |
2018 8th IEEE International Conference on Control System, Computing and Engineering (ICCSCE). |
| DOI: |
10.1109/iccsce.2018.8685010 |
| Popis: |
Audio (speech) is part of multimedia. Audio has its own properties. Therefore, analysis audio is an important aspect of multimedia signal processing. Speech rate and frequency are two important aspects of characterizing expressive speech. Speech rate is the speed of continuously spoken words. The rate of speech in words per minute from slow to fast, with normal rate averaging about 125 words per minutes. In the frequency measures, the range of human speech varies from 80 to 260 Hertz. Both features define the quality of speech especially in upholding comfort among listeners. In this study, speeches from seven prominent speakers have been collected randomly and analyzed. Semi-automated transcription has been completed. The speech rate and voice frequency were computed, and its range has been classified. Results and findings show that each speaker has a different speech rate and frequency. As each speaker have a high volume of followers, we recommended their speech characters as the range of ideal rate of 140wpm – 160wpm for the speech rate and range of 125Hz – 250Hz for the voice frequency. |
| Databáze: |
OpenAIRE |
| Externí odkaz: |
|
