Download presentation
Presentation is loading. Please wait.
1
Speech Communications Chapter 7
2
Speech Communications The Nature of Speech Criteria for Evaluating Speech Components of Speech Communication System Synthesized Speech
3
The Nature of Speech 1/2 發聲 : 呼吸系統, Articulators Types of Speech Sound Phoneme ( 音素 ) − shortest segment of speech if change → meaning change if change → meaning change 分類 : 母音 (vowel), 子音 (consonant) 雙母音 (diphthongs) 雙母音 (diphthongs) Phoneme →Syllable →Word → Sentence
4
The Nature of Speech 2/2 Depicting Speech Waveform, Spectrum Sound spectrogram Fig 8-1 Fig 8-1 Fig 8-1 Intensity of Speech Average intensity (speech power): 母音>子音 Intelligibility: 子音較重要 Frequency Composition of Speech 低頻 : 男>女 Fig 8-2 Fig 8-2 Fig 8-2 Shouting: frequency 上升
5
Criteria for Evaluating Speech Speech Intelligibility ( 能解度 ) 方法 − Repeat 呈現的聲音 − 回答問題 Test − Nonsense syllables − Isolated words (phonetically balanced, PB) − Sentences Speech quality (Naturalness) Preference
6
Components of Speech Communication System Speaker Message Transmission System Noise Hearer
7
Components of Speech Communication System Speaker Enunciation ( 清晰的聲音 ) Superior Speakers − Longer syllable duration − Greater intensity − More total time with speech sounds − Frequencies varied 1/7
8
Components of Speech Communication System Message Phoneme Confusion − DVPBGCET, FXSH, KJA, MN − Avoid single letters, Word-spelling alphabet Word Characteristics − Familiar words − Long words 2/7
9
Components of Speech Communication System Message Context Features − Sentence: meaningful > nonsense − Set size: 字多<字少 Fig 7-3 Fig 7-3 Fig 7-3 − Guidelines 用較少的字 Standard sentence Avoid short word Familiarize user 3/7
10
Components of Speech Communication System Transmission System Filtering (Frequency distortion) Fig 7-4 Fig 7-4 Fig 7-4 − High-pass: cutoff < 600 Hz − Low-pass: cutoff > 4000 Hz Amplitude Distortion Fig 7-5 7-6 Fig 7-57-6 Fig 7-57-6 − Peak clipping Quality , Intelligibility ≈ − Center clipping Intelligibility − 提高 Intelligibility: Peak clipping Amplify ( 子音 / 母音 ) 4/7
11
Components of Speech Communication System Noise Articulation Index (AI) Fig 7-7 Fig 7-7 Fig 7-7 − 1/3 octave, S-N, weighted sum − Intelligibility Fig 7-8 Tab 7-1 Fig 7-8Tab 7-1 Fig 7-8Tab 7-1 Preferred-Octave Speech Interference Level (PSIL) − Mean of 500, 1000, 2000 Hz (octave) − SIL: Mean of 600-1200, 1200-2400,... − Intelligibility (vs. distance) Fig 7-9 Fig 7-9 Fig 7-9 − Subjective rating Fig 7-10 Tab 7-2 Fig 7-10Tab 7-2 Fig 7-10Tab 7-25/7
12
Components of Speech Communication System Noise Preferred Noise Criterion Curve (PNC) Fig 7-11Fig 7-11 Tab 7-3 Tab 7-3 Fig 7-11Tab 7-3 Reverberation Fig 7-12 Fig 7-12 Fig 7-12 − Reverberation time: Decay 60 dB − Reverberation time Intelligibility 6/7
13
Components of Speech Communication System Hearer Age Fig 7-13 Fig 7-13 Fig 7-13 Wearing of Hearing Protection 7/7
14
Synthesized Speech 種類 Uses Performance Preference Guidelines
15
Synthesized Speech 種類 Synthesis by Analysis − Digitized human speech compressed data format compressed data format − 缺點 : 限於 encoded & stored Lack of coarticulation Lack of coarticulation Synthesis by Rule − 缺點 : quality 較差
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.