--- title: "Speech characteristics in coder design" subject: "Wireless Communication" chapter: 6 key: "semester_7/wireless_communication:speech-characteristics-coder-design" source: https://www.bctengineer.com/q/semester_7/wireless_communication/speech-characteristics-coder-design site: BCT Engineering --- # Speech characteristics in coder design Wireless Communication · Chapter 6: Speech and Channel Coding - Asked in: 2082 Bhadra, 4 marks; 2081 Baisakh, 4 marks; 2072 Asoj, 8 marks (asked 3 times) - Answer checked by Claude Opus 5.5 (high effort) on 2026-10-14 - Page: https://www.bctengineer.com/q/semester_7/wireless_communication/speech-characteristics-coder-design - Whole chapter: https://www.bctengineer.com/md/semester_7/wireless_communication/chapter-6 ## Question What are different characteristics of speech signals? How are they used in designing of coders? ## Short answer Speech signals have limited 300-3400 Hz bandwidth, non-uniform pdf, high sample correlation, non-flat spectrum with formants, pitch periodicity, short-term stationarity over 20 ms frames, and silence periods. Coders exploit these by using 8 kHz sampling, non-uniform quantization, predictive coding (DPCM/ADPCM), LPC modeling, sub-band coding, voice activity detection, and perceptual weighting. ## Answer #### Characteristics of speech signals 1. **Limited bandwidth** – intelligible speech lies mostly in 300–3400 Hz. 2. **Non-uniform probability density** – low amplitudes occur far more often than high ones (Laplacian/gamma-like pdf). 3. **Non-zero autocorrelation** – adjacent samples (125 µs apart at 8 kHz) are highly correlated (correlation coefficient often > 0.85). 4. **Non-flat spectrum** – energy is concentrated at low frequencies, with resonant peaks called **formants**. 5. **Quasi-periodicity (pitch)** – voiced sounds repeat with a pitch period of about 3–12 ms (80–350 Hz); unvoiced sounds are noise-like. 6. **Short-term stationarity** – statistics stay nearly constant over 10–30 ms segments. 7. **Silence periods** – speakers are silent about 40–60% of the time. 8. **Perceptual masking and phase insensitivity** – the ear is relatively insensitive to phase and to noise near strong spectral components. #### How these are used in coder design | Characteristic | Use in coder design | |---|---| | Bandwidth 300–3400 Hz | Sample at 8 kHz; filter out other bands | | Non-uniform pdf | Non-uniform quantization (μ-law/A-law) gives more levels to small amplitudes | | Sample correlation | Predictive coding (DPCM, ADPCM) sends only the prediction error, fewer bits | | Non-flat spectrum | Sub-band and adaptive transform coding give more bits to strong bands | | Formants | LPC models the vocal tract as an all-pole filter; only ~10 coefficients sent | | Pitch periodicity | Long-term (pitch) predictor; voiced excitation = pulse train | | Short-term stationarity | Process in 20 ms frames; update parameters once per frame | | Silence | Voice activity detection and discontinuous transmission save power and capacity | | Perceptual masking | Perceptual weighting filter in CELP shapes noise under the speech spectrum | **Example:** the GSM full-rate RPE-LTP coder uses 20 ms frames, an 8th-order short-term LPC filter (formants) and a long-term predictor (pitch) to reach 13 kbps instead of 64 kbps PCM. ## Check yourself The answer's key points as multiple-choice questions, with the correct option and why. 1. Which coding technique exploits the fact that low speech amplitudes are far more probable? A. Sub-band coding B. Non-uniform quantization (mu-law/A-law) C. Pitch prediction D. Voice activity detection Answer: B. Giving more quantization levels to small amplitudes matches the Laplacian-like pdf. (Key point: Non-uniform pdf use) 2. Which coding approach uses the high correlation between adjacent samples? A. Predictive coding such as DPCM and ADPCM, sending the prediction error B. Companding C. Frame processing D. Discontinuous transmission Answer: A. Prediction error has a smaller range, so fewer bits are needed. (Key point: Correlation use) 3. How does LPC use the formant structure of speech? A. It removes silence B. It models the vocal tract as an all-pole filter and sends only about 10 coefficients C. It samples at 16 kHz D. It filters out 300-3400 Hz Answer: B. Formants are vocal-tract resonances, well modelled by an all-pole filter with few coefficients. (Key point: Formants and LPC) 4. Which characteristic leads to voice activity detection and discontinuous transmission? A. Pitch periodicity B. Perceptual masking C. Non-flat spectrum D. Speakers are silent 40-60% of the time Answer: D. Silence periods allow transmission to be switched off, saving power and capacity. (Key point: Silence use) 5. What is the use of perceptual masking in CELP? A. Pitch detection B. Frame length selection C. A perceptual weighting filter shapes noise under the speech spectrum D. Bandwidth reduction to 300-3400 Hz Answer: C. The ear tolerates noise near strong spectral components, so noise is shaped to be masked. (Key point: Perceptual masking use) 6. The GSM full-rate RPE-LTP coder reduces the bit rate from 64 kbps PCM to about what? A. 2.4 kbps B. 13 kbps C. 32 kbps D. 26 kbps Answer: B. RPE-LTP with LPC and a long-term predictor reaches 13 kbps. (Key point: GSM rate) ## Related in Wireless Communication - [Source coders used in speech coding](https://www.bctengineer.com/question.md?key=semester_7%2Fwireless_communication%3Aspeech-source-coders) — Wireless Communication, chapter 6 - [Speech characteristics and linear predictive coder](https://www.bctengineer.com/question.md?key=semester_7%2Fwireless_communication%3Aspeech-lpc) — Wireless Communication, chapter 6, asked 5× - [Speech characteristics; formant and channel vocoder](https://www.bctengineer.com/question.md?key=semester_7%2Fwireless_communication%3Aspeech-formant-channel-vocoder) — Wireless Communication, chapter 6 - [Speech characteristics and 7-bit Hamming code](https://www.bctengineer.com/question.md?key=semester_7%2Fwireless_communication%3Aspeech-hamming-code) — Wireless Communication, chapter 6 - [Characteristics of speech signal](https://www.bctengineer.com/question.md?key=semester_7%2Fwireless_communication%3Aspeech-characteristics) — Wireless Communication, chapter 6 - [Formant vocoder and speech characteristics](https://www.bctengineer.com/question.md?key=semester_7%2Fwireless_communication%3Aformant-vocoder-speech) — Wireless Communication, chapter 6 ## Related in other subjects - [Characteristics and signals of 8085](https://www.bctengineer.com/question.md?key=semester_4%2Fmicroprocessor%3A8085-characteristics-signals) — Microprocessor, chapter 2