08-20-1227, chapter 4
This commit is contained in:
@@ -214,16 +214,16 @@ export default withMermaid(
|
||||
link: "/sounds-of-american-english/4-natural-speech",
|
||||
items: [
|
||||
{
|
||||
text: "4.1. 句子",
|
||||
link: "/sounds-of-american-english/4.1-sentences",
|
||||
text: "4.1. 音节",
|
||||
link: "/sounds-of-american-english/4.1-syllables",
|
||||
},
|
||||
{
|
||||
text: "4.2. 词汇",
|
||||
link: "/sounds-of-american-english/4.2-words",
|
||||
},
|
||||
{
|
||||
text: "4.3. 音节",
|
||||
link: "/sounds-of-american-english/4.3-syllables",
|
||||
text: "4.3. 句子",
|
||||
link: "/sounds-of-american-english/4.3-sentences",
|
||||
},
|
||||
],
|
||||
},
|
||||
|
||||
Binary file not shown.
Binary file not shown.
BIN
1000-hours/public/audios/marm1.wav
Normal file
BIN
1000-hours/public/audios/marm1.wav
Normal file
Binary file not shown.
BIN
1000-hours/public/audios/marm2.wav
Normal file
BIN
1000-hours/public/audios/marm2.wav
Normal file
Binary file not shown.
BIN
1000-hours/public/audios/marm3.wav
Normal file
BIN
1000-hours/public/audios/marm3.wav
Normal file
Binary file not shown.
BIN
1000-hours/public/audios/marm4.wav
Normal file
BIN
1000-hours/public/audios/marm4.wav
Normal file
Binary file not shown.
@@ -11,12 +11,14 @@
|
||||
|
||||
在**自然语流**中,每个**音素**都可能存在一定的变化。
|
||||
|
||||
任何一个音素都一样,实际上并不存在一个像音乐音符那样可以 100% 精确的标准。时时刻刻,每个音素都有**长短**、**强弱**、**高低**、**起伏**、**轻重**、**缓急**等等各个维度上并不统一的变化 —— 不仅如此,说话的每个人又有着各自的特质,包括但不限于无法一致的音质、音域、音量、语速、腔调、情绪,等等等等…… 也正因如此,最终每个人的说话方式都各不相同 —— 实际上是没办法完全相同。其实,不仅英语如此,地球上的所有语言都是如此。
|
||||
任何一个音素都一样,实际上并不存在一个像音乐音符那样可以 100% 精确的标准。时时刻刻,每个**音素**(或其组合,**音节**)都可能有**长短**、**强弱**、**高低**、**起伏**、**轻重**、**缓急**等等各个维度上并不统一的变化。
|
||||
|
||||
不仅如此,说话的每个人又有着各自的特质,包括但不限于无法一致的音质、音域、音量、语速、腔调、情绪,等等等等…… 也正因如此,最终每个人的说话方式都各不相同 —— 实际上是没办法完全相同。其实,不仅英语如此,地球上的所有语言都是如此。
|
||||
|
||||
即便是相同的单词,在同一句话里都常常读法并不完全相同,也无法完全相同 —— 注意两个相同的词的每个音节的**音高**和**声调**的不同:
|
||||
|
||||
> * **communication**: Her *communication* skills are excellent, but her *communication* of the project details needs work.<span class="speak-word-inline" data-audio-us-male="/audios/Her-communication-skills-are-excellent-but-her-communication-of-the-project-details-needs-work-alloy.mp3" data-audio-us-female="/audios/Her-communication-skills-are-excellent-but-her-communication-of-the-project-details-needs-work-nova.mp3"></span>
|
||||
> * **explanation**: The *explanation* you gave was clear, but I need a more detailed *explanation*.<span class="speak-word-inline" data-audio-us-male="/audios/The-explanation-you-gave-was-clear-but-I-need-a-more-detailed-explanation-alloy.mp3" data-audio-us-female="/audios/The-explanation-you-gave-was-clear-but-I-need-a-more-detailed-explanation-nova.mp3"></span>
|
||||
> * **communication** <span class="pho alt">kəˌmjuː.nəˈkeɪ.ʃən</span>: Her *communication* skills are excellent, but her *communication* of the project details needs work.<span class="speak-word-inline" data-audio-us-male="/audios/Her-communication-skills-are-excellent-but-her-communication-of-the-project-details-needs-work-alloy.mp3" data-audio-us-female="/audios/Her-communication-skills-are-excellent-but-her-communication-of-the-project-details-needs-work-nova.mp3"></span>
|
||||
> * **explanation** <span class="pho alt">ˌek.spləˈneɪ.ʃən</span>: The *explanation* you gave was clear, but I need a more detailed *explanation*.<span class="speak-word-inline" data-audio-us-male="/audios/The-explanation-you-gave-was-clear-but-I-need-a-more-detailed-explanation-alloy.mp3" data-audio-us-female="/audios/The-explanation-you-gave-was-clear-but-I-need-a-more-detailed-explanation-nova.mp3"></span>
|
||||
|
||||
1974 年,美国密西根州立大学(Michigan State University)的人工语言实验室的研究人员曾经打电话用机器生成的语音订购一块披萨[^1]…… 以下是机器语音合成 50 年前后的对比:
|
||||
|
||||
@@ -27,7 +29,7 @@
|
||||
> * OpenAI TTS (Alloy)<span class="speak-word-inline" data-audio-us-male="/audios/Would-you-please-phrase-that-question-so-that-I-can-answer-it-with-yes-or-no_openai.mp3"></span>
|
||||
> * Microsoft Edge TTS (en-US-GuyNeural)<span class="speak-word-inline" data-audio-us-male="/audios/Would-you-please-phrase-that-question-so-that-I-can-answer-it-with-yes-or-no_msedge.mp3"></span>
|
||||
|
||||
显然,模拟真人的自然语流,并不只是 “把每个音素朗读标准” —— 除此之外需要考虑的因素实在是太多,而各个维度的不同再组合起来就是天文数字的量级…… 乃至于需要将近 50 年的时间,以计算机算力的提高、算法的改良进步为前提,而后还要配合着大规模神经网络以及基于大语言模型的人工智能才有了这样极其接近 “真实”。
|
||||
显然,模拟真人的自然语流,并不只是 “把每个音素朗读标准” —— 除此之外需要考虑的因素实在是太多,而各个维度的不同再组合起来就是天文数字的量级…… 乃至于需要将近 50 年的时间,以计算机算力的提高、算法的改良进步为前提,而后还要配合着大规模神经网络以及基于大语言模型的人工智能才有了如此极其接近 “真实” 的效果。
|
||||
|
||||
我们用自己的嗓音说话也是如此。要做的事情,不仅仅是 “把每个音素” 读准读好,也不仅仅是 “把每个单词读得像词典里的真人发音一样”,我们需要从多个维度调整自己 —— 当然很麻烦,不过,事实证明,也的确是能做到做好的事情。
|
||||
|
||||
|
||||
@@ -1,8 +0,0 @@
|
||||
# 4.1 句子
|
||||
|
||||
一句话由一个或者多个词汇构成 —— 当一句话由多个词汇构成的时候,有个重要的现象:
|
||||
|
||||
> * 有些词会被强读(*accented form*)
|
||||
> * 有些词会被弱读(*weak form*)
|
||||
|
||||
即,不是每个词都是以同样的方式读出来的。
|
||||
133
1000-hours/sounds-of-american-english/4.3-sentences.md
Normal file
133
1000-hours/sounds-of-american-english/4.3-sentences.md
Normal file
@@ -0,0 +1,133 @@
|
||||
# 4.1 句子
|
||||
|
||||
一句话由一个或者多个词汇构成 —— 当一句话由多个词汇构成的时候,有个重要的现象:
|
||||
|
||||
> * 有些词会被**强读**(***accented** form*)
|
||||
> * 有些词会被**弱读**(***weak** form*)
|
||||
|
||||
即,不是每个词都是以同样的方式读出来的。
|
||||
|
||||
仔细听听以下例句:
|
||||
|
||||
> **TEXT**:
|
||||
> * The art of focus in our whirlwind existence can sometimes feel like searching for a needle in a haystack.
|
||||
> * The <span class="pho alt">ðɪ</span> art <span class="pho alt">ˈɑːrt</span> of <span class="pho alt">ɒv</span> focus <span class="pho alt">ˈfoʊkəs</span> in <span class="pho alt">ɪn</span> our <span class="pho alt">ˌaʊɚ</span> whirlwind <span class="pho alt">ˈwɝːlwɪnd</span> existence <span class="pho alt">ˈɪɡzɪstəns</span> can <span class="pho alt">kæn</span> sometimes <span class="pho alt">ˈsʌmtaɪmz</span> feel <span class="pho alt">ˈfiːl</span> like <span class="pho alt">ˈlaɪk</span> searching <span class="pho alt">ˈsɝːtʃɪŋ</span> for <span class="pho alt">fɔːr</span> a <span class="pho alt">ə</span> needle <span class="pho alt">ˈniːdəl</span> in <span class="pho alt">ɪn</span> a <span class="pho alt">ə</span> haystack <span class="pho alt">ˈheɪstæk</span>
|
||||
>
|
||||
> **AUDIO**:
|
||||
> * 逐字朗读版本<span class="speak-word-inline" data-audio-us-male="/audios/The-art-of-focus-in-our-whirlwind-existence-can-sometimes-feel-like-searching-for-a-needle-in-a-haystack-all-strong.mp3"></span>
|
||||
> * 自然连贯版本 <span class="speak-word-inline" data-audio-us-male="/audios/The-art-of-focus-in-our-whirlwind-existence-can-sometimes-feel-like-searching-for-a-needle-in-a-haystack-natural.mp3"></span>
|
||||
> * 弱读的词汇:The art *of* focus *in* *our* whirlwind existence *can* sometimes feel *like* searching *for* a needle *in* a haystack.
|
||||
> * 相对强读的词汇:The **art** of **focus** in our **whirlwind** existence can **sometimes** **feel** like **searching** for a **needle** in a **haystack**.
|
||||
|
||||
这是一上来就需要调整(或者纠正)的习惯:
|
||||
|
||||
> **不能像词典一样读句子里的每一个单词**。
|
||||
|
||||
## 4.1.1. 弱读词汇
|
||||
|
||||
首先要学会的是,把一些 “虚词” —— 根据情况,比如助动词、系动词、介词、连词、冠词、代词等等 —— 弱读。
|
||||
|
||||
比如,在以上的例句中<span class="speak-word-inline" data-audio-us-male="/audios/The-art-of-focus-in-our-whirlwind-existence-can-sometimes-feel-like-searching-for-a-needle-in-a-haystack-natural.mp3"></span>,*of* 不是 <span class="pho alt">ɒv</span> 而是 <span class="pho alt">əv, əf, ə</span> 甚至 <span class="pho alt">ɤ</span>,*in* 不是带着重音的 <span class="pho alt">ˈɪn</span> 而是很轻的 <span class="pho alt">ɪn, ən</span>, *our* 不是 <span class="pho alt">ˌaʊɚ</span> 而是 <span class="pho alt">ɑr</span>,*can* 不是 <span class="pho alt">kæn</span> 而是 <span class="pho alt">kən</span>,*like* <span class="pho alt">laɪk</span> 中的双元音 <span class="pho">aɪ</span> 读得很短,*for* 不是 <span class="pho alt">fɔːr</span> 而是 <span class="pho alt">fɚ</span>……
|
||||
|
||||
以下是其中最常见的强读式、弱读式对照列表。
|
||||
|
||||
> [!Note]
|
||||
> 这个列表不能当作**硬规则**使用,不是所有的虚词都必须被弱读;也不是所有的实词都必须被强读 —— 根据情况,这些词在自然语流中也有被强读的时候。
|
||||
|
||||
> * a: <span class="pho alt">eɪ</span> → <span class="pho alt">ə</span>
|
||||
> * am: <span class="pho alt">æm</span> → <span class="pho alt">əm, m</span>
|
||||
> * an: <span class="pho alt">æn</span> → <span class="pho alt">ən, n</span>
|
||||
> * and: <span class="pho alt">ænd</span> → <span class="pho alt">ənd, nd, ən, n</span>
|
||||
> * any: <span class="pho alt">'eni</span> → <span class="pho alt">ni</span>
|
||||
> * are: <span class="pho">/ɑ/:</span> → <span class="pho alt">ə</span>
|
||||
> * as: <span class="pho alt">æs</span> → <span class="pho alt">əz</span>
|
||||
> * at: <span class="pho alt">æt</span> → <span class="pho alt">ət</span>
|
||||
> * but: <span class="pho alt">bʌt</span> → <span class="pho alt">bət</span>
|
||||
> * can: <span class="pho alt">kæn</span> → <span class="pho alt">kən, kn, kŋ</span>
|
||||
> * could: <span class="pho alt">kud</span> → <span class="pho alt">kəd, kd</span>
|
||||
> * do: <span class="pho alt">duː</span> → <span class="pho alt">du, də, d</span>
|
||||
> * does: <span class="pho alt">dʌz</span> → <span class="pho alt">dəz</span>
|
||||
> * for: <span class="pho alt">fɔː</span> → <span class="pho alt">fə, f</span>
|
||||
> * from: <span class="pho alt">frɔm</span> → <span class="pho alt">frəm, frm</span>
|
||||
> * had: <span class="pho alt">hæd</span> → <span class="pho alt">həd, əd, d</span>
|
||||
> * has: <span class="pho alt">hæz</span> → <span class="pho alt">həz, əz</span>
|
||||
> * have: <span class="pho alt">hæv</span> → <span class="pho alt">həv, həf, əv, əf</span>
|
||||
> * he: <span class="pho alt">hiː</span> → <span class="pho alt">hi, iː, i</span>
|
||||
> * her: <span class="pho alt">həː</span> → <span class="pho alt">hə, əː, ə</span>
|
||||
> * him: <span class="pho alt">him</span> → <span class="pho alt">im</span>
|
||||
> * his: <span class="pho alt">hiz</span> → <span class="pho alt">iz</span>
|
||||
> * I: <span class="pho alt">ai</span> → <span class="pho alt">aː, ə</span>
|
||||
> * is: <span class="pho alt">iz</span> → <span class="pho alt">s, z</span>
|
||||
> * many: <span class="pho alt">'meni</span> → <span class="pho alt">mni</span>
|
||||
> * me: <span class="pho alt">miː</span> → <span class="pho alt">mi</span>
|
||||
> * must: <span class="pho alt">mʌst</span> → <span class="pho alt">məst, məs</span>
|
||||
> * my: <span class="pho alt">mai</span> → <span class="pho alt">mi</span>
|
||||
> * of: <span class="pho alt">əv</span> → <span class="pho alt">əv, v, f, ə</span>
|
||||
> * our: <span class="pho alt">ɑʊɚ</span> → <span class="pho">/ɑr</span> ⭢ <span class="pho">ɑ/</span>
|
||||
> * shall: <span class="pho alt">ʃæl</span> → <span class="pho alt">ʃəl, ʃl</span>
|
||||
> * she: <span class="pho alt">ʃiː</span> → <span class="pho alt">ʃi</span>
|
||||
> * should: <span class="pho alt">ʃud</span> → <span class="pho alt">ʃəd, ʃd, ʃt</span>
|
||||
> * so: <span class="pho alt">səʊ</span> → <span class="pho alt">sə</span>
|
||||
> * some: <span class="pho alt">sʌm</span> → <span class="pho alt">səm, sm</span>
|
||||
> * such: <span class="pho alt">sʌʧ</span> → <span class="pho alt">səʧ</span>
|
||||
> * than: <span class="pho alt">ðæn</span> → <span class="pho alt">ðən, ðn</span>
|
||||
> * that: <span class="pho alt">ðæt</span> → <span class="pho alt">ðət</span>
|
||||
> * the: <span class="pho">/ði/:</span> → <span class="pho alt">ði, ðə</span>
|
||||
> * them: <span class="pho alt">ðem</span> → <span class="pho alt">ðəm, ðm, əm, m</span>
|
||||
> * then: <span class="pho alt">ðen</span> → <span class="pho alt">ðən</span>
|
||||
> * to: <span class="pho alt">tuː</span> → <span class="pho alt">tu, tə, t</span>
|
||||
> * us: <span class="pho alt">us</span> → <span class="pho alt">əs</span>
|
||||
> * was: <span class="pho alt">wɔz</span> → <span class="pho alt">wəz, wə</span>
|
||||
> * we: <span class="pho alt">wiː</span> → <span class="pho alt">wi</span>
|
||||
> * were: <span class="pho alt">wəː</span> → <span class="pho alt">wə</span>
|
||||
> * when: <span class="pho alt">wen</span> → <span class="pho alt">wən</span>
|
||||
> * will: <span class="pho alt">wɪl</span> → <span class="pho">/əl, l</span>/
|
||||
> * would: <span class="pho alt">wuːd</span> → <span class="pho alt">wəd, əd</span>
|
||||
> * you: <span class="pho alt">juː</span> → <span class="pho alt">ju, jə</span>
|
||||
|
||||
## 4.1.2. 强读词汇
|
||||
|
||||
事实上,在自然语流中,讲话者经常调整强读(强调)的词汇,用来表达不同的含义。试比较以下同一个句子的 4 个版本[^1],在不同的版本里,被强读的词汇不同:
|
||||
|
||||
> * **Marianna** made the **marmalade**.<span class="speak-word-inline" data-audio-us-female="/audios/marm1.wav"></span>
|
||||
> * **Marianna** made the marmalade.<span class="speak-word-inline" data-audio-us-female="/audios/marm2.wav"></span>
|
||||
> * **Marianna** **made** the **marmalade**.<span class="speak-word-inline" data-audio-us-female="/audios/marm3.wav"></span>
|
||||
> * Marianna made the **marmalade**?<span class="speak-word-inline" data-audio-us-female="/audios/marm4.wav"></span>
|
||||
|
||||
拿第 1 个版本<span class="speak-word-inline" data-audio-us-female="/audios/marm1.wav"></span>为例,请参照以下图示:
|
||||
|
||||

|
||||
|
||||
同样一个句子,**强调**不同的**词**可能表达不同的含义[^2]:
|
||||
|
||||
> * Will **you** drive to the office tommorrow? (rather than someone else)
|
||||
> * Will you **drive** to the office tommorrow? (rather than going some other way)
|
||||
> * Will you drive **to** the office tommorrow? (rather than from the office)
|
||||
> * Will you drive to the **office** tommorrow? (rather than somewhere else)
|
||||
> * Will you drive to the office **tommorrow**? (rather than some other day)
|
||||
|
||||
被**强读**的词汇之中,重音音节往往带着**声调** —— 升、降、平、降升。并且,少数重音音节会有被拔高的**音高**。这些都是自然语流的韵律中格外重要的组成部分,也是跟读模仿练习时的重点之一。
|
||||
|
||||
## 4.1.3. 强弱带来的变化
|
||||
|
||||
如果一个音节是重音或被**强读**,那么这个单词中的:
|
||||
|
||||
> * 长元音会被读的很清楚,并且足够长,甚至显得更长一些;
|
||||
> * 双元音会被读的很饱满,并且显得很有弹性;
|
||||
> * 落在重音上的短元音都会显得更长一些;
|
||||
> * 重音所在的音节可能带着不同的声调(平调、升调、降调、降升)……
|
||||
> * 重音所在的音节可能带着更高的音高(往往是 “高”、“中”、“低” 中的 “高”)……
|
||||
|
||||
如果一个音节被**弱读**,那么这个单词中的:
|
||||
|
||||
> * 长元音会变得短一些(几乎与短元音的长度相当);
|
||||
> * 重音音节会变得与非重音音节一样轻;
|
||||
> * 很多元音都会发生变化,向 *schwa* <span class="pho">ə</span>靠拢;
|
||||
> * 很多与辅音相关的停顿会被直接省略;
|
||||
> * 清辅音 <span class="pho">s</span>、<span class="pho">t</span>、<span class="pho">k</span>、<span class="pho">f</span> 之后的元音 <span class="pho">ə</span> 可能会直接被省略掉;
|
||||
> * 整个单词所处的音高往往是 “高”、“中”、“低” 之中的 “低”,最多是 “中”……
|
||||
|
||||
其实,哪怕在单独读某一个单词的时候,其中元音的长短都会受到重音重读的影响。比如,*city* 这个单词,重音在第一个音节上,而两处的元音是一样的:<span class="pho alt">ˈsɪ-t̬ɪ</span>;但是只要你把第一个音节读得足够重,自然而然就能感觉到**第一个音节**只能比**第二个音节**更长。
|
||||
|
||||
[^1] 这个例子来自于 [Macquarie University](https://www.mq.edu.au/about/about-the-university/our-faculties/medicine-and-health-sciences/departments-and-centres/department-of-linguistics/our-research/phonetics-and-phonology/speech/phonetics-and-phonology/Intonation-tobi-introduction)
|
||||
[^2] 这个例子来自于 *Manual of American English Pronunciation*, 4th edition, by Clifford H. Prator, Jr. BettyWallaceRobinett, 1972
|
||||
Reference in New Issue
Block a user