Unit Selection Speech Synthesis Using Phonetic-Prosodic Description of
Speech Databases
Tetyana Lyudovyk, Mykola Sazhok
International Research/Training Center for Information Technologies and Systems, Kyiv, Ukraine
{tetyana_lyudovyk, mykola}@https://www.doczj.com/doc/db11836455.html,.ua
Abstract
This paper describes an approach to speech synthesis based on using speech databases at different stages of TTS process. Speech database units are phones in different segmental and prosodic contexts. Pitch synchronous segmentation and labeling of databases allows storing both segmental and prosodic information.
Phonetic-prosodic annotations of speech databases are involved in off-line training of the linguistic processor. The automatic transcriptor, duration and intonation modules are trained to model the speech characteristics of different persons and thus to generate different target specifications of one and the same input text during the synthesis stage. A target specification is a detailed phonetic-prosodic transcription used by the unit selection module.
The unit selection algorithm is based on criteria derived from categories of phonetic-prosodic annotations of speech databases and works without spectral matching. The output of the unit selection module is an acoustic phonetic-prosodic transcription which is used by the acoustic processor to generate a speech wave.
Two non-professional speaker databases with different speaking styles have been created and tested. 1.Introduction
Speech databases play a great role in concatenative synthesizers. Extending a database size and coverage we increase the probability of finding speech units with specified properties, e.g context, duration and F0 contour. Consequently we decrease the need to modify the speech signal. As a result we obtain the synthesized speech which is more natural and of higher quality.
Speech databases are individual and reflect the characteristic pronunciation properties of speakers. Using this information ensures that a speaker will be recognized by the synthesized speech. The novelty of the suggested approach consists in using the database phonetic-prosodic annotation not only for on-line unit selection but also for off-line training of all modules of the linguistic processor. The trained linguistic processor is able to generate speaker-specific target specifications of input texts. The main idea is that the closer the target specification is to the database speaker pronunciation, the more precise and correct unit selection and overall synthesis quality will be achieved.
The overall approach is time domain oriented. It concerns database collection, segmentation and labeling, as well as unit selection criteria used and signal processing techniques applied (optional).
The described approach is realized in the experimental Ukrainian TTS system. Two speech databases with male voices (8000 and 12000 units) are developed and a speech database with a female voice is under developement.
https://www.doczj.com/doc/db11836455.html,ponents of the Ukrainian TTS system
The experimental Ukrainian TTS system is composed of the traditional modules [1], [2], [3]: speech database, linguistic processor, acoustic processor, and unit selection module [3], [4], [5], which
is indispensable in synthesizers using large speech databases. The overall system architecture is shown in Figure 1.
Speech database plays the main role. The information it contains (detailed annotation of acoustic units corresponding to phones in context) is used by all other modules of the synthesizer.
The pre-tuning is applied offline to all modules of the linguistic processor (text normalization, accentuation and phrasing, phonetic transcription, assignment of duration and F0 contour.
The annotation of the speech database is used also as the main information source for unit selection during the synthesis stage. The criteria used by the unit selection algorithm are based on the annotation categories: left and right context of a unit, its duration and F0, as well as the sequence of units in the speech corpus preserved in the database.
Speech signal segments stored in the database are used by the acoustic processor for generating the resulting speech waveform corresponding to the input text [6].
3.Speech database
The quality of the synthesized speech depends on the size and the coverage of the speech database. Another important factor is the level of detail of the description which is used to represent the speech units.
3.1. Database design
Speech database units are phones (phonemes and distinct allophones) in context. Phones have been chosen because they are the main segmental phonetic units and appear as main elements of target specifications generated for input texts.
We distinguish 58 phones in Ukrainian: 6 stressed and 6 unstressed vowels, 22 non-palatalized consonants, 23 palatalized consonants, and pause. So, stress is present explicitly in the phoneset, which is typically the case for Eastern Slavic language synthesizers [1], [2].
Different speech corpora are recorded using non-professional speakers reading aloud texts and isolated sentences. Recordings are made at 22 kHz sampling rate.
Multiple instances of each phone differing in duration and pitch are stored.
Phones are collected in the order they occur in training sentences. This information is used by unit selection algorithm instead of continuity distance measurement.
The size of speech corpora used for database creation is moderate, which allows to manually correct
its labeling and segmentation. As a result the annotations of speech databases are reliable and reflect the slightest peculiarities of speaker pronunciation. This fact allows for using these speech databases not only for text-to-speech synthesis, but also for creation of linguistic models (e.g. prosody model) and for different kinds of experimental-phonetic research.
3.2. Database collection and segmentation
The collection of a database is done under expert control with possibility to see and hear every unit instance thanks to interactive visualization tools [7]. The segmentation of speech corpus into phones is done manually, because it requires correcting of automatically obtained transcription of training texts [8] to how a particular speaker reads these training texts (different accentuation, presence/absence of assimilation, reduction, insertion of extra pauses etc.). The second stage of segmentation, into pitch periods, is carried out automatically [6], [9], but it also requires an expert supervision.
Pitch marks are positioned on zero crossing points before the first negative or positive peak of the pitch period, depending on speech material. Each pitch period is visually checked and possibly corrected. This
Figure 1. Block diagram of the Ukrainian TTS system
allows to obtain the accurate description of F0 movement, reflecting the details of microprosody.
The database is pruned by removing the majority of pauses, as well as “poorly” pronounced phones.
3.3. Database unit description
Each phone in context may occur several times in the database, so untis need identifiers. Segmentation provides information about unit duration and F0 trajectory. Only voiced units are segmented into pitch periods.
A fragment of one annotation is shown in Figure 2. Each instance of a unit in the database is labeled with: ?unit instance identifier;
?three-part unit name (preceding, current, succeeding phone names);
?unit instance duration in ms;
?number of pitch periods (only for voiced units); ?average pitch period length (only for voiced units); ?first pitch period length (only for voiced units); ?medium pitch period length (only for voiced units); ?last pitch period length (only for voiced units).
2725 a - l - Y 64.08 10 6.39 5.90 6.62 6.17
2726 l - Y - s 80.32 15 5.35 6.08 5.31 5.03
2727 Y - s - a 115.51
2728 s - a - h 68.21 11 6.17 5.53 6.17 6.94
2729 a - h - o 83.95 11 7.62 7.39 7.53 7.71
2730 h - o - l 55.42 7 7.89 7.53 7.98 8.30
2731 o - l - o 40.18 5 8.03 7.89 8.07 8.12
2732 l - o - v 62.54 8 7.80 6.98 8.89 8.16
2733 o - v - A 61.63 8 7.66 8.07 7.62 7.30
2734 v - A - # 159.41 27 5.90 7.17 5.90 4.90
Figure 2. A fragment of the speech database annotation.
As a result, a database unit description comprises both segmental and prosodic characteristics of speech. 4.Linguistic processor training
The goal of the linguistic processor in a unit selection concatenative speech synthesizer is to generate the target specification of an input orthographic text. Typically, linguistic processors are independent of the database [3], [5], [10], [11]. Our goal, on the contrary, is to produce such a target specification that would model a speaker specific pronunciation (e.g. different word accentuation, different pitch range, different pitch contours for continuation rise, enumeration etc.). We believe that making target specification close to database pronunciation would help significantly with unit selection as well as with naturalness and quality of the synthesized speech.
Linguistic processor components use two types of data: speaker independent and speaker dependent. Speaker independent data are the phoneset and phones features. Speaker dependent data are:
?dictionaries for text normalization and word stress
marking;
?maximum length of an intonation group (in phonetic words);
?rules for phonetic assimilation and reduction; ?average durations of phones and duration lengthening and shortening coefficients;
?intonation contour inventories.
Let us examine some of the speaker dependent data taking as an example two speech databases with male voices. The maximum length of an intonation group is correspondingly 7 and 8 phonetic words. The average durations of stressed vowels are:
?[A]: 117 ms and 95 ms;
?[O]: 102 ms and 84 ms;
?[E]: 104 ms and 84 ms;
?[I]: 88 ms and 68 ms;
? [Y]: 87 ms and 69 ms;
? [U]: 94 ms and 78 ms.
The same speakers use different F0 contours for non-finality. The fundamental frequency assignment module produces speaker dependent target contours, as shown in Figure 3.
Figure 3. Speaker dependent target F0 curves in case of the same intonation group with non-finality (in transliteration): “Zajshovshy do banku“ (“entered the bank”).
At present the fully automated training is developed for phone duration prediction. The phone duration prediction is based on the phone average duration and a set of duration coefficients. Each phone of each speaker has its own average duration and its own set of duration coefficients. Moreover, these data depend on the speaking style. For synthesizing speech in different speaking styles we suggest to use different speech
databases recorded from one donor-speaker.
5.Target specification
The target specification provided by the linguistic processor is the phonetic-prosodic transcription of the input text. To obtain the target specification the following procedures are involved: text normalization, and segmentation into intonation groups and intonation type assignment, word stress marking, grapheme-to-phone conversion, segmenting of intonation groups into accent groups, computing F0 curves and duration of phones, conversion of the sequence of F0 values into the sequence of pitch period lengths. The resulting target phonetic-prosodic transcription is used during unit selection.
The target phonetic-prosodic transcription for the same intonation group “Zajshovshy do banku“ is present in Figure 4. Phones-in-context are followed by sequences of pitch period lengths for voiced phones and simply by duration for unvoiced phones (in ms).
#-#-z 50
#-z-a 8.0 8.0 8.0 8.0 8.0 8.0 8.0 8.0 8.0 8.0 8.0 8.0
z-a-j 8.0 8.0 7.9 7.9 7.9 7.8 7.8 7.8 7.7
a-j-sh 7.7 7.7 7.7 7.7 7.7 7.7 7.7 7.7
j-sh-O 80
sh-O-v 7.4 7.4 7.3 7.3 7.2 7.2 7.2 7.2 7.3 7.3 7.4 7.5 7.5 7.6 7.7
O-v-sh 7.7 7.7 7.7 7.7 7.7 7.7 7.7 7.7
v-sh-y 80
sh-y-d 7.7 7.7 7.7 7.7 7.7 7.7 7.7 7.7 7.7 7.7
y-d-o 7.7 7.7 7.7 7.7 7.7 7.7 7.7 7.7 7.7 7.7 7.7 7.7
d-o-b 7.7 7.7 7.6 7.6 7.6 7.6 7.5 7.5 7.5 7.4 7.4
o-b-A 7.4 7.4 7.4 7.4 7.4 7.4 7.4 7.4 7.4 7.4 7.4 7.4 7.4 7.4
b-A-n 7.4 7.3 7.3 7.2 7.1 7.1 7.0 7.0 6.9 6.8 6.6 6.5 6.4 6.4 6.3 6.6
6.9
7.3 7.4 7.5 7.7
A-n-k 7.7 7.7 7.7 7.7 7.7 7.7 7.7 7.7
n-k-u 70
k-u-# 8.3 8.3 8.3 8.3 8.3 8.3 8.3 8.3 8.3 8.3 8.3 8.3 8.3 8.3
u-#-# 80
Figure 4. Target specification for the intonation group “Zajshovshy do banku”.
6.Unit selection
The unit selection algorithm is based on phonetic and prosodic criteria. It uses:
?target phonetic-prosodic transcription of the input text provided by the linguistic processor; ?phonetic-prosodic annotation of the speech database;
?tables of phonetic-acoustic distances between Ukrainian phones;
?phonetically motivated criteria of selection.
The main criterion of unit selection algorithm according to phoneme-triphone model [12] is contextual identity of target and candidate units. Left and right neighbors of each target unit are taken into account. The algorithm first searches the database for candidate units that match the target unit along with its left and right contexts. If there are such units in the database, the selection continues in two different ways for periodic and non-periodic units.
For periodic units the following prosodic selection criteria are used:
?difference between target and candidate units in average pitch period lengths;
?difference between target and candidate units in pitch periods number.
For non-periodic phones the difference between target and candidate units durations is used as the selection criterion.
One more criterion is the immediate vicinity of units in the database. The selection of consecutive phones is encouraged.
If the database does not contain units with left and right segmental context equal to the target segmental context, tables that specify the phonetic-acoustic distances between Ukrainian phones are used to search for unit with context similar to the target unit context.
The result of unit selection procedure is a specification for the acoustic processor. This specification is called acoustic phonetic-prosodic transcription and it differs from the target phonetic-prosodic transcription by:
?the indication of database units identifiers;
?the indication for periodic units what pitch periods in what quantity (for unit duration changing) and with what length (for unit F0 changing) should be taken for the concatenation.
The resulting unit selection specification for the intonation group presented in sections 4 and 5 is shown in Figure 5. Note that for several units some lengthening or shortening is required (certain pitch periods must be repeated or deleted). Units neighboring in the database are presented in bold. Database units with context not identical to the target unit context are presented in italic.
1271 #-#-z 0 50
2162 #-z-a (1 8.0) (2 8.0) (4 8.0) (5 8.0) (6 8.0)
(7 8.0)
(9 8.0)(10 8.0)(11 8.0)(12 8.0)(14 8.0)(15 8.0) 4240 z-a-j (1 8.0)(2 8.0)(4 7.9)(5 7.9)(7 7.9)(8 7.8)
(10 7.8)(11 7.8)(13 7.7)
4164 a-j-sh (1 7.7)(2 7.7)(3 7.7)(3 7.7)(4 7.7)(5 7.7)
(6 7.7)(6 7.7)
4165 j-sh-O 0 80
4166 sh-O-v (1 7.4)(2 7.4)(4 7.3)(5 7.3)(7 7.2)(8 7.2)
(10 7.2)(11 7.2)(13 7.3)(14 7.3)(16 7.4)(17 7.5)
(19 7.5)(20 7.6)(22 7.7)
1004 O-v-t(1 7.7)(2 7.7)(2 7.7)(3 7.7)(4 7.7)(4 7.7)
(5 7.7)(5 7.7)
4759 v-sh-y 0 80
2065 k-y-d(1 7.7)(2 7.7)(2 7.7)(3 7.7)(4 7.7)
(4 7.7)(5 7.7)(6 7.7)(7 7.7)(7 7.7)
2590 y-d-o (1 7.7)(2 7.7)(3 7.7)(3 7.7)(4 7.7)(5 7.7)
(6 7.7)(7 7.7)(8 7.7)(8 7.7)(9 7.7)(10 7.7)
974 d-o-d(1 7.7)(2 7.7)(3 7.6)(4 7.6)(5 7.6)
(6 7.6)(7 7.5)(8 7.5)(9 7.5)(10 7.4)(11 7.4) 2156 o-b-A (1 7.4)(2 7.4)(2 7.4)(3 7.4)(4 7.4)
(5 7.4)(5 7.4)(6 7.4)(7 7.4)(7 7.4)(8 7.4)(9 7.4)
(10 7.4)(10 7.4)
5594 b-A-n (1 7.4)(2 7.3)(3 7.3)(4 7.2)(5 7.1)(6 7.1)
(7 7.0)(8 7.0)(9 6.9)(10 6.8)(11 6.6)(11 6.5)
(12 6.4)(13 6.4)(14 6.3)(15 6.6)(16 6.9)(17 7.3)
(18 7.4)(19 7.5)(20 7.7)
5595 A-n-k (1 7.7)(2 7.7)(4 7.7)(5 7.7)(7 7.7)(8 7.7)
(9 7.7)(11 7.7)
3140 n-k-u 0 70
1168 k-u-# (1 8.3)(2 8.3)(3 8.3)(4 8.3)(5 8.3)
(6 8.3)(7 8.3)(7 8.3)(8 8.3)(9 8.3)(10 8.3)
(11 8.3)(12 8.3)(13 8.3)
539 u-#-# 0 80
Figure 5: Unit selection module output for the intonation group “Zajshovshy do banku”.
7.Unit concatenation
After the appropriate sequence of units is selected from the database, the next step is to concatenate the segments. This is done by the acoustic processor.
To obtain the prosody specified in the target, it is usually necessary to modify the duration and fundamental frequency.
Our time domain approach ensures prosodic modification of selected units via signal processing. The prosodic modification of a database unit consists in repeating or omitting its pitch periods according to the target acoustic phonetic-prosodic transcription, as well as in lengthening or shortening of selected pitch periods applying the linear prediction model [12].
The speech signal processing inevitably distortes the real speech segments and thus degrades the naturalness of synthesized speech. The size of the speech database is important, because among a great number of segmentally and contextually identical phones it is more probably to find a phone with appropriate prosodic characteristics.
At present two versions of speech generation are available. The first version modifies each selected pitch period of selected unit according to target specification. The second version does not modify selected units at all; that is, they are taken from the database with their real duration and pitch.
Consistent pitch marking of speech data and detailed target specification of prosody allow to avoid phase and pitch mismatches at speech generation stage [13], while taking into account coarticulation reflected in contexts allows to minimize spectral mismatch.
8.Results
Three allophone-based databases with three different donor speakers have been created:
?artificial words with careful monotone rather slow pronunciation (712 items);
?real speech, sentences from stories and novels (12057 items);
?real speech, radio-news reports (8433 items).
A listening test was conducted to evaluate, how the speech recordings quality and the signal processing which modifies the prosody influence the intelligibility and the naturalness of synthesized speech.
Two texts differing in style (a news report and a fiction story) were synthesized using the first and the second databases under following combinations:
A)with small database, without unit selection,
with prosodic modification;
B)with medium size database (5873 items), with
unit selection, with prosodic modification;
C)with medium size database (5873 items), with
unit selection, without prosodic modification.
20 listeners, all native speakers of Ukrainian from different regions of Ukraine were asked to rate A, B, and C versions of synthesized speech according to intelligibility and naturalness.
The results are shown in Figure 6.
Intelligibility
Naturalness
Figure 6: Listeners’ preference percentage for three tested synthesis technique combinations A, B, C.
The results of listening test indicate that the best intelligibility is achieved with small database. This is due to correct delivery of speech data during recordings. As for naturalness, the version C is preferred. This indicates that listeners tolerate spectral or prosodic mismatches if the quality of the synthetic speech closer approaches that of natural speech. Version B keeps an intermediate position.
A preliminary experiment related to fully automated segmentation and labeling of speech data was carried out with female speech recordings.
9.Future work
The future work will be concentrated on: ?selecting optimal speech data for databases; ?automatic creation of speech databases;
?pruning the size of databases by removing unit instances which are redundant or are never selected
due to their values out of established pitch or duration range;
?providing better representations for target, adding amplitude (energy) specification;
?automatic training of all modules of linguistic processor;
?refining unit selection criteria;
?elaborating new signal processing techniques; ?trying non-uniform units (demi-phones, phone clusters etc.).
10.Conclusions
The main goal of the work described was to develop and test a new speech technology based on:
?careful database design, segmentation and annotation;
?producing target specification close to database content in segmental and prosodic sense;
?simple unit selection algorithm.
Concatenative synthesis approach is able to retain the characteristics of the donor speaker in the synthesized speech. Speaker characteristics are reflected in the target specification.
Detailed segmental-prosodic annotation of database units and detailed phonetic-prosodic target specification allow selection criteria to be simple and easily manipulated.
The listening test showed that an important factor influencing the intelligibility of resulting synthesized speech is the database quality [14]. As for the naturalness, the strategy “select the best, modify the least” is confirmed to be true.
11.References
[1]Krivnova, O.F. “Automatic synthesis of Russian
speech”, Proceedings of the XIV International
Congress of Phonetic Sciences, Vol.1, 507–510,
San Francisco, 1999.
[2]ЛобановБ.М., КарневскаяЕ.Б., ЛевковскаяТ.В.
“Синтезаторречипотекстукаккомпьютерное
средство?клонирования? персонального
голоса”, Тр. Международнойконференции
Диалог-2001, 265-272, М., 2001.
[3]Beutnagel, M., Conkie A., Schroeter, J., Stylianou,
Y. and Syrdal, A. “The AT&T Next-Gen TTS
System”, The Proceedings of the Joint Meeting of
ASA, EAA, and DAGA, 18-24, Berlin, Germany,
1999.
[4]Hunt, A., Black, A. W. “Unit selection in a
concatenative speech synthesis system using a large
speech database”, Proceedings of the IEEE International Conference on Acoustics and Speech
Processing, Vol. 1, 373-376, Munchen, Germany,
1996.
[5]Coorman, G., Fackrell, J., Rutten, P., Van Coile,
B. “Segment Selection in the L&H RealSpeak
Laboratory TTS System”, Proceedings of the
International Conference on Spoken Language
Processing, Vol. 2, 395-398, Beijing, China,2000. [6]ЛюдовикТ.В., СажокН.Н. “Использование
речевыхбазданныхбольшогообъемапри
синтезеречивсистемахискусственного
интеллекта”, Проблемыуправленияи
информатики, 6, 82-87, 2003.
[7]Sazhok, M. “Speech modelling virtual laboratory”,
Proc eedings of the International Summer School on
Speech Processing, Recognition and Artificial
Networks, 229–233, Vietri-sul-Mare, Italy, 1998. [8]Vintsiuk, T. K., Liudovyk, T.V., Sazhok, M.M.,
“Phonetic Knowledge Base for Ukrainian”, Proceedings of the Second International Conference SPECOM’98, 179-192, St.Petersburg,
Russia, 1998.
[9]Vintsiuk, T. K., “Optimal joint procedure for
current pitch period discrimination and speech
signal partition into quasi-periodic and non-
periodic segments”, Proceedings of the First International Workshop on Text, Speech, Dialogue—TSD’98, 135-140, Brno, 1998.
[10]Conkie, A. “Robust unit selection system for
speech synthesis”, The Proceedings of the Joint
Meeting of ASA, EAA, and DAGA, 18-24, Berlin,
Germany, 1999.
[11]Beutnagel, M., Conkie, A., and Syrdal, A.
“Diphone synthesis using unit selection”, The
Proceedings of the 3rd ESCA/COCOSDA International Workshop on Speech Synthesis, 185-
190, Jenolan Caves, Australia, 1998.
[12]ТарасВ?нцюк, ТетянаЛюдовик, МиколаСажок,
РусланСелюх. “Автоматичнийозвучувач
укра?нськихтекст?внаоснов?фонемно-
трифонно?модел?звикористаннямприродного
мовногосигналу”, Прац? 6-?Всеукра?нсько?
м?жнародно?конференц??“Обробленнясигнал?в
?зображеньтарозп?знаванняобраз?в” – УкрОбраз’2002, Ки?в, 2002, с. 79–84.
[13]Bozkurt, B., Bagein, M., Dutoit, T. “From
MBROLA to NU-MBROLA”, Proceedings of the
4th ISCA Speech Synthesis Workshop, 127-129,
Pitlochry, Scotland, 2001.
[14]M?bius B. “Corpus-based speech synthesis:
methods and challenges”, Arbeitspapiere des Institut fur Maschinelle Sprachverarbeitung,AIMS
6(4), 87-116, University of Stuttgart, 2000.
作文翻译 Unit 1 李明是学化学的，性格开朗幽默，颇有魅力，但英语成绩不佳，每次只能勉强及格。老师警告他，英语不好会阻碍他拿奖学金，并亮出了自己的王牌：如果李明不努力，就让他考试不过关。老师还告诉他，学习英语不能只为了文凭，否则他即使大学毕业，也还是个半文盲。李明虽然保持镇定，但他明白，他的学业生涯正在攸关之际，必须安心下来埋头学习，坚持不懈。 Li Ming was a chemistry major, a charmer noted for his easygoing and humorous temperament. However, his English was so poor that he always barely got by. The teacher admonished him that his poor English would be an impediment to scholarship. What’s more, she showed her trump card: if Li Ming did not work hard. She would flunk him. He was also told that he should not learn English merely for the sake of his diploma. otherwise, even after graduation from university, he would still be semiliterate. Although Li Ming did not lose his composure, he was well aware that he had to settle down to work and follow through because his academic life was at stake. Unit2 我的朋友琳达接受过良好的教育，既美丽又端庄，三十好几依然没有人向她求婚。究其原因，她的事业心极强，整日扑在工作上，每天来往于住处和公司之间，根本没有时间和异性交往。一想到女儿这么大了还单身一人，她父母就焦虑不安。他们不知道该如何是好，甚至还去咨询一些社会学专家。 但是事情在上个月出现了转机，公司的总部调琳达到培训部。在新的工作岗位上，琳达遇到了第一个触动她心弦的男人。从此，他们几乎每天约会，琳达意识到她会不顾一切地爱这个男人。决定嫁人的时候，她告诉了我这个好消息。 虽然琳达的爱情让人想起电影中才会有的浪漫故事，我也担忧未来究竟会怎样，但我还是表达了我由衷的祝福，并爽快答应在婚礼那天做他们的伴娘和伴郎随从中的一员。 Linda, my good friend, has received good education and is both beautiful and elegant. She was not proposed to even when she was well over thirty. The reason is that she, as a career –oriented woman, is devoted to her work. Navigating between home and the company, she had hardly any time to socialize with people of the opposite sex. Her parents were gripped by anxiety at the thought of their daughter still remaining single at such an age. They did not know what to do and even consulted with some sociologists. But the situation began to change last month, when the headquarters of the company transferred Linda to the training department. On the new post, Linda met a man who tugged on her heartstrings for the first time. Ever since then, they dated virtually on a daily basis, and Linda realized that she would love the man beyond all reason. When she decided to take the matrimonial plunge, she informed me. Though Linda’s love is reminiscent of the romance that we see only in movies and I don’t know what the future will hold for her, I give her my heart-felt wishes and agree readily to be a member of the entourage of bridesmaids and groomsmen. Unit 3 食品供应商缺乏诚信已经成为当今社会的一大问题。部分企业欺骗公众，故意散布假消息，颂扬食品添加剂是食品工业的伟大成就，并声称适量的添加剂对健康有益无害。部分有良知的科学家对食品添加剂的含量和毒性展开了深入的病理学研究。研究结果表明，部分常见的食品添加剂经长期，可能会对健康产生危害，这被认为是食品安全研究方面极为重要的
个人公司章程 公司章程 一、总则 第一条依据《中华人民共和国公司法》、《中华人民共和国公司登记管理条例》及有关法律、法规的规定，制定本公司章程。本公司章程对公司股东、董事、监事、经理均具有约束力。 第二条公司经公司登记机关核准登记并领取法人营业执照后即告成立。 二、公司名称和住所 第三条公司名称：有限公司。(以预先核准登记的名称为准) 第四条公司住所：市(县镇)路(街)号。 三、公司的经营范围 第五条公司的经营范围：(含经营方式)。 四、公司注册资本 第六条公司的注册资本为全体股东实缴的出资总额，人民币万元。(要符合法定的注册资本的最低限额) 第七条公司注册资本的增加或减少必须经股东会代表2/3以上表决权股东一致通过，增加或减少的比例、幅度必须符合国家有关法律、法规的规定，而且不应影响公司的存在。 五、公司股东名称
第八条凡持有本公司出具的认缴出资证明的为本公司股东，股东是法人的，由该法人的法定代表人或法人的代理人代表法人行使股东权利。 第九条公司在册股东共人，全部是法人股东 股东名录： (一)法人股东： 1.法人名称： 住所： 法定代表人： 认缴出资额：万元，占公司注册资本的 % 出资方式： (货币或实物或其它) 认缴时间：年月日 2.…………………………………… 第十条公司置备股东名册，并记载下列事项： (一)股东的姓名或者名称及住所; (二)股东的出资额; (三)出资证明书编号。 六、股东的权利和义务 第十一条公司股东享有以下权利： 1.出席股东会，按出资比例行使表决权;
2.按出资比例分取公司红利; 3.有权查询公司章程、股东会会议记录、财务会计报表; 4.公司新增资本时，可优先认缴出资; 5.按规定转让出资; 6.其它股东转让出资，在同等条件下，有优先购买权; 7.有权在公司解散清算时按出资比例分配剩余财产; 第十二条公司股东承担以下义务： 1.遵守公司章程; 2.按期缴足认购的出资; 3.以其出资额为限对公司承担责任; 4.出资额只能按规定转让，不得退资; 5.有责任保护公司的合法权益，不得参与危害公司利益的活动; 6.在公司登记后，不得抽回出资; 7.在公司成立后，发现作为出资的实物、工业产权、非专利技术、土地使用权的实际价额显著低于公司章程所定价额的，应当由交付该出资的股东补交其差额，公司设立时的其他股东对其承担连带责任 七、股东(出资人)的出资方式和出资额 第十三条出资人以货币认缴出资额。(以实物、工业产权、
企业先进个人事迹 【总厂之星·铿锵玫瑰】记xxx单位xxx同志 勇于开始才能找到成功的路 --记xxx单位xxx同志 xxx单位xxx xxx，一名默默奉献于总厂数字化业务的技术尖兵。2006年研究生毕业后xxx来到xxxxxx 工作，曾先后在抽油机制造分厂、产品研发中心从事技术研发。自总厂业务转型开始，她又积极投身其中，凭着自己勤于学习、善于钻研、勇于实践的性格，用短短三年时间，在数字化新产品、新技术开发应用方面结出了丰硕成果，成为xxxxxx数字化业务转型的铺路石，数字化技术创新的火炬手，数字化技术应用的传道者，数字化抽油机推广的护航人。 冲锋一线数字化业务转型的铺路石 2009年，xxxxxx按照xxxxxx要求制定了以数字化为核心推进业务转型的发展思路，并从全厂抽调了二十名技术人员探索涉入这一全新的业务领域。xxx是这二十名技术人员里唯一的女性，但作为总厂唯一的自动化专业的研究生，她又必须承担为总厂数字化业务转型闯出一条新路的重托，她没有犹豫，毅然放弃了研发中心优裕的办公环境和新婚不久的家庭，同其他男同志一道上前线学习、钻研数字化新技术，她充分运用专业所长，在团队里交流学习心得，帮促了其他技术人员的迅速成长，也自然而然地成为了这支技术团队的核心和带头人。现场施工时，她又与男同志一道爬电杆、装设备、接线、调试，亲力亲为、奋战一线，直到一天在去井场的颠簸道路上，她蜷缩在车后排的一角，脸上的汗珠直往下掉，才被大家送往医院检查，原来她早有身孕，大家都埋怨她不该隐瞒情况，她只歉意地解释现在是施工的关键期她不想离开。 刻苦钻研数字化技术创新的火炬手 根据xxxxxx "数字化关键技术要为xxx自己掌握"的指示精神，xxxxxx成立了专门的数字化产品研发团队，xxx同志以扎实的专业功底和实践能力成为了这支研发团队的副组长。按照"立足自主研发，技术研发与技术引进相结合"的原则，她带领研发组一边收集资料，学习国内外成熟的数字化技术与经验，一边拖着日渐笨重的身体，跑高校、院所和专业公司，谈技术引进与合作，顺利签署了多项合作协议。同时，她和团队也完成了 "数字化采油系统故障检测与现场维护系统开发"等8项xxxxxx和总厂重点科研项目，先后获得xxxxxx科学技术进步奖一等奖和二等奖各一项，"从式井场套管自动加药装置"等四项技术获得的股份公司实用新型专利授权，2010年她所在团队又荣获总厂"建功立业"先进集体称号，她也荣获女职工"建功立业"先进个人称号。 无私奉献数字化技术应用的传道者 通过刻苦钻研学习，xxx同志掌握了扎实的数字化技术功底，但她追求的不是一支独秀，而是满园芬芳。在xxxxxx数字化业务纵深推进过程中，她在人员培训以及数字化新技术、新产品现场推广应用方面倾注了大量心血。2011年，xxxxxx数字化技术服务在油田范围内全面铺开，因为服务大队是整建制转型， 前线人员的技术水平参差不齐，关于各种技术问题的求助电话铺天盖地打到技术研发团队。一方面，xxx耐心地对各种问题给予解答或远程技术指导，另一方面，她迅速组织技术团队编写了《数字化井站维护服务操作指南》，按照实际操作程序和步骤重点介绍了井场油系统、阀组间、视频监控系统等不同厂家和型号设备的配置、调试方法，为现场维护服务工作的顺利开展提供了强有力的技术支撑和保证。 勇挑重担数字化抽油机推广的护航人 第二代数字化抽油机在油田大范围推广应用，xxx作为数字化抽油机控制系统子项目的负责人，她主动承担了该项核心技术的现场技术支持工作，起草编制了《第二代数字化抽油
Unit1 The Burden of Womanhood 女人的负担 Too often in the Third World, a female?s life is hardly worth living. 往往在第三世界，女性的生活几乎是不值得活的。 By: John Ward Anderson & Molly Moore 约翰.沃德.安德森和莫莉.摩尔1.When Rani returned home from the hospital cradling her newborn daughter, the men in the family slipped out of her mud hut while she and her mother-in-law mashed poisonous oleander seeds into a dollop of oil and dropped it into the baby?s throat. As soon as darkness fell, Rani crept into a nearby field and buried her baby girl in a shallow, unmarked grave next to a small stream. 当拉尼抱着她刚出生的女儿从医院回到家时，男人们溜出了她的土屋，她和她的婆婆将有毒的夹竹桃种子捣碎成一团油，并将其塞到婴儿的喉咙里。夜幕一降临，拉尼就蹑手蹑脚地来到附近的一块土地，将她的女婴埋在一条小溪浅旁边的一个不深的、没有标记的坟墓内。 2.“I never felt any sorrow,” Rani, a farm laborer with a weather-beaten face, said through an interpreter. “There was a lot of bitterness in my heart toward the baby because the gods should have given me a son.” “我从来没有感到任何悲哀”，脸部饱经风霜的拉尼通过翻译说，“我的心对孩子充满了辛酸，因为神本应该赐予我一个儿子”。 3.Each year hundreds and perhaps thousands of newborn girls in India are murdered by their mothers simply because they are female. Some women believe that sacrificing a daughter guarantees a son in the next pregnancy. In other cases, the family cannot afford the dowry that would eventually be demanded for a girl?s marriage. 在印度每年有数百甚至数千名新生女婴被他们的母亲谋杀，只因为他们是女孩。有些妇女认为，牺牲一个女儿将保证下次怀孕是儿子。在其他情况下，家庭买不起女孩结婚时所需的嫁妆。 4.And for many mothers, sentencing a daughter to death is better than condemning her to life as a woman in the Third World, with cradle-to-grave discrimination, poverty, sickness and drudgery. 对于很多母亲而言，判处女儿死刑胜过让她在第三世界作为一个女人生活，经历一辈子的歧视、贫困、疾病和劳苦。 5. While women in the United States and Europe-----after decades of struggling for equal rights----often measure sex discrimination by pay scales and seats in corporate board rooms, women in the Third World gauge discrimination by mortality rates and poverty levels. 虽然欧美国家的妇女，在经过了几十年对平等权利的抗争之后，通常以工薪标准和董事会中的席位来衡量性别歧视，但第三世界的妇女以死亡率和贫困程度来衡量歧视。 6. “Women are the most exploited among the oppressed,” says Kauna Chanana Ah med, a New Delhi anthropologist who has studied the role of women in developing countries. “I don?t think it ?s even possible to eradicate discrimination, it?s so deeply ingrained.” “妇女是最受压迫和剥削的”，研究妇女在发展中国家的地位的新德里的人类学家Karuna
个人开公司必备：注册、税务 一、选择公司的形式：- 普通的有限责任公司，最低注册资金3万元，需要2个（或以上）股东，从06年1月起新的公司法规定，允许1个股东注册有限责任公司，这种特殊的有限责任公司又称“一人有限公司”（但公司名称中不会有“一人”字样，执照上会注明“自然人独资”），最低注册资金10万元。-- 如果你和朋友、家人合伙投资创业，可选择普通的有限公司，最低注册资金3万元；如果只有你一个人作为股东，则选择一人有限公司，最低注册资金10万元。- 二、注册公司的步骤：- 1.核名：到工商局去领取一张“企业（字号）名称预先核准申请表”，填写你准备取的公司名称，由工商局上网（工商局内部网）检索是否有重名，如果没有重名，就可以使用这个名称，就会核发一张“企业（字号）名称预先核准通知书”。这一步的手续费是30元。（30元可以帮你检索5个名字，很多名字重复，所以一般常见的名字就不用试了，免得花冤枉钱） 2.租房：去专门的写字楼租一间办公室，如果你自己有厂房或者办公室也可以，有的地方不允许在居民楼里办公。租房后要签订租房合同，并让房东提供房产证的复印件。- 签订好租房合同后，还要到税务局去买印花税，按年租金的千分之一的税率购买，例如你的每年房租是1万元，那就要买10元钱的印花税，贴在房租合同的首页，后面凡是需要用到房租合同的地方，都需要是贴了印花税的合同复印件。-- 3.编写“公司章程”：可以在工商局网站下载“公司章程”的样本，修改一下就可以了。章程的最后由所有股东签名。- 4.刻私章：去街上刻章的地方刻一个私章，给他们讲刻法人私章（方形的）。费用大概20元左右。-
5.到会计师事务所领取“银行询征函”：联系一家会计师事务所，领取一张“银行询征函”（必须是原件，会计师事务所盖鲜章）。如果你不清楚，可以看报纸上的分类广告，有很多会计师事务所的广告。-- 6.去银行开立公司验资户：所有股东带上自己入股的那一部分钱到银行，带上公司章程、工商局发的核名通知、法人代表的私章、身份证、用于验资的钱、空白询征函表格，到银行去开立公司帐户，你要告诉银行是开验资户。开立好公司帐户后，各个股东按自己出资额向公司帐户中存入相应的钱。- 银行会发给每个股东缴款单、并在询征函上盖银行的章。- 注意：公司法规定，注册公司时，投资人（股东）必须缴纳足额的资本，可以以贷币形式（也就是人民币）出资，也可以以实物（如汽车）、房产、知识产权等出资。到银行办的只是货币出资这一部分，如果你有实物、房产等作为出资的，需要到会计师事务所鉴定其价值后再以其实际价值出资，比较麻烦，因此建议你直接拿钱来出资，公司法不管你用什么手段拿的钱，自己的也好、借的也好，只要如数缴足出资款即可。- 7.办理验资报告：拿着银行出具的股东缴款单、银行盖章后的询征函，以及公司章程、核名通知、房租合同、房产证复印件，到会计师事务所办理验资报告。一般费用500元左右（50万以下注册资金）。- 8.注册公司：到工商局领取公司设立登记的各种表格，包括设立登记申请表、股东（发起人）名单、董事经理监理情况、法人代表登记表、指定代表或委托代理人登记表。填好后，连同核名通知、公司章程、房租合同、房产证复印件、验资报告一起交给工商局。大概3个工作日后可领取执照。此项费用约300元左右。- 9.凭营业执照，到公安局指定的刻章社，去刻公章、财务章。后面步骤中，均需要用到公章或财务章。--
公司先进岗位事迹材料 锐意创新，建立行业领先的销售管理平台 ××集团公司销售部经理××自××年加入本公司以来，在领导的关心、支持和同事的配合下，比较圆满地完成了自己所承担的各项工作任务，在政治思想觉悟和业务工作能力等方面都取得了较大的进步，为今后的工作和学习打下了良好的基础。爱思范文23: 该同志政治立场坚定，遵纪守法；工作勤奋敬业，积极进取，团结协作，恪守职业道德，遵守社会公德；求真务实，作风严谨，诚实守信，具有良好的学术道德。 该同志在工作中勤于学习，具有扎实的专业理论基础，致力创新。经过销售部市场主管岗位及目前销售部经理岗位上五年多的工作，通过不断学习、不断积累，该同志具备了比较丰富的工作经验，能够比较从容地处理日常工作中出现的各类问题，在组织管理能力、综合分析能力、协调办事能力和文字言语表达能力等方面，都有了很大的提高，不但保证了本岗位各项工作的正常运行，而且能够较好地把握业务发展方向，结合业务工作提出新办法，建立健全销售管理平台和管理体系，极大的提高了本公司销售管理水平，使公司的销售管理在制药行业中保持领先。 一、建立crm系统，实现销售管理网络化、信息化、制度化。在以往传统的销售管理体系中，普遍存在销售过程难以监控、人员行为难以控制、经销商行为无法监管、销售费用难以控制、销售决策因无准确的市场信息所造成的主观失误等问题。针对以上的问题，该同志根据管理学的进步和信息发展的趋势，在公司领导的支持和同事的配合下，建立了利用现代信息与网络技术，将互联网与集团局域网融为一体的crm销售管理系统。crm销售管理系统的核心工作是围绕公司的市场、销售人员、经销商通过有效的方式准确地反映销售人员的工作状况，销售费用的发生状况，产品的销售状况及市场信息的变化状况等动态的销售信息。整个销售管理系统的监控考核的核心是市场和销售人员。在这些准确的销售动态信息的基础之上形成相关的统计分析数据，既是对市场及销售人员进行考核的量化指标又是提供给销售决策者重要的基本参照物。依托该系统，本公司以消费者满意为核心，建立起消费者→药店→二级经销商→一级经销商→集团总部的商业网络，并在此基础上建立起业务员、医务人员网络，从药店到患者，建立起家庭健康档案，履行家庭医师的职责，服务于患者。crm系统通过现代化的手段，解决了一系列销售管理的难题，为公司销售费用的控制及销售业绩的大幅提升做出了贡献，极大地提高了企业的营销运行效率，大大降低了企业的经营成本。
翻译 Unit 1 李明是学化学的，性格开朗幽默，颇有魅力，但英语成绩不佳，每次只能勉强及格。老师警告他，英语不好会阻碍他拿奖学金，并亮出了自己的王牌：如果李明不努力，就让他考试不过关。老师还告诉他，学习英语不能只为了文凭，否则他即使大学毕业，也还是个半文盲。李明虽然保持镇定，但他明白，他的学业生涯正在攸关之际，必须安心下来埋头学习，坚持不懈。 Li Ming was a chemistry major, a charmer noted for his easygoing and humorous temperament . However, his English was so poor that he always barely got by. The teacher admonished him that his poor English would be an impediment to scholarship. What’s more, she showed her trump card: if Li Ming did not work hard. She would flunk him. He was also to ld that he should not learn English merely for the sake of his diploma. otherwise, even after graduation from university, he would still be semiliterate. Although Li Ming did not lose his composure, he was well aware that he had to settle down to work and follow t hrough because his academic life was at stake. Unit2 我的朋友琳达接受过良好的教育，既美丽又端庄，三十好几依然没有人向她求婚。究其原因，她的事业心极强，整日扑在工作上，每天来往于住处和公司之间，根本没有时间和异性交往。一想到女儿这么大了还单身一人，她父母就焦虑不安。他们不知道该如何是好，甚至还去咨询一些社会学专家。但是事情在上个月出现了转机，公司的总部调琳达到培训部。在新的工作岗位上，琳达遇到了第一个触动她心弦的男人。从此，他们几乎每天约会，琳达意识到她会不顾一切地爱这个男人。决定嫁人的时候，她告诉了我这个好消息。虽然琳达的爱情让人想起电影中才会有的浪漫故事，我也担忧未来究竟会怎样，但我还是表达了我由衷的祝福，并爽快答应在婚礼那天做他们的伴娘和伴郎随从中的一员。 Linda, my good friend, has received good education and is both beautiful and elegant. She was not proposed to even when she was well over thirty. The reason is that she, as a career -oriented woman, is devoted to her work. Navigating between home and the company, she had hardly any time to socialize with people of the opposite sex. Her parents were gripped by anxiety at the thought of their daughter still remaining single at such an age. They did not know what to do and even consulted with some sociologists. But the situation began to change last month, when the headquarters of the company transferred Linda to the training department. On the new post, Linda met a man who tugged on her heartstrings for the first time. Ever since then, they dated virtually on a daily basis, and Linda realized that she would love the man beyond all reason. When she decided to take the matrimonial plunge, she informed me.Though Linda’s love is reminiscent of the romance that we see only in movies and I don’t know what the future will hold for her, I give her my heart-felt wishes and agree readily to be a member of the entourage of bridesmaids and groomsmen. Unit3食品供应商缺乏诚信已经成为当今社会的一大问题。部分企业欺骗公众，故意散布假消息，颂扬食品添加剂是食品工业的伟大成就，并声称适量的添加剂对健康有益无害。部分有良知的科学家对食品添加剂的含量和毒性展开了深入的病理学研究。研究结果表明，部分常见的食品添加剂经长期，可能会对健康产生危害，这被认为是食品安全研究方面极为重要
个人注册公司应该选择哪种类型的个人注册公司应该选择哪种类型的，对于初次创业的人来说，在资金、社会关系、业务渠道、经营实力等方面处于弱势，初次创业的人要经过详细的了解和计算来选择公司的组织形式，尽量选择最适合自己的公司形式注册。下面雁荡上海注册公司，对适合个人注册的几个类型进行了简要介绍和优劣对比分析。一、个人独资企业 个人独资企业是指由一个自然人投资,财产为投资者个人所有,并以其个人财产对企业债务承担无限责任的经营实体。 （一）个人独资企业其优势主要体现在: 1、设立门槛低。没有法定最低注册资本金的要求,投资人只需有申报的出资即可。 2、企业自由度大,经营管理灵活。个人独资企业财产归投资人所有,由投资人支配;投资人可以自行管理,也可以选择聘任管理或委托管理。 3、企业负税轻。个人独资企业非企业所得税纳税主体,企业所得不必缴纳企业所得税,投资人只须就个人投资所得交纳个人所得税。 （二）个人独资企业其劣势主要体现在： 个人独资企业作为非法人企业,不能以自己的财产独立承担有限责任,所以个人独资企业一旦资不抵债,投资人应以自己的全部财产偿还企业不能偿还的债务,承担的风险较大。 二、一人有限责任公司 一人有限责任公司简称一人公司，它是由一个自然人或者法人投资设立的有限责任公司。相对于个人独资企业而言，尽管投资人都是一人，但二者却存在根本的不同。一人公司是法人企业，有自己独立支配的法人财产，公司财产与投资人的个人财产严格分开，投资人以认缴的出资为限对公司承担责任，公司以自己所有的财产对债权人承担有限责任。 （一）一人公司的劣势主要体现在： 1、一人公司设立门槛比较高;公司法规定，注册一人公司，最低资本限额十万元人民币，而且必须一次缴足。 2、投资人负税较重。公司必须就其所得缴纳企业所得税，投资人从公司取
Unit 1> Caring for Our Earth Passage A （三）、 1.Answer: A tree frog has taken up residence in my studio. Answer: He stopped and turned around and just sat there looking at me. 2.Answer: Possibly the tone of my computer sounded to him like other tree frogs. 3.Answer: Frogs were dying around the world. Answer: Frogs were being found whose skin was like paper. Answer: Frogs are an “indicator species”. 4.Answer: The frogs have a message for us. Answer: We must be the adults for the planet. Answer: There are no boundaries, that there is no more time. （四）、Answer: greenhouse residence communicate wired humming message sensitivity indicator （五）、Answer: 1. d 2. a 3. b 4. e 5. c 1. tone threatening 2. sensitivity smell 3. affected climate 4. tap resources 5. concerns environment （六）、Answer: 1. subtle 2. were overtaken 3. species 4. decades 5. boundaries 6. audio 7. directly 8. focus 9. tone 10. cupped （七）、Answer: 1. at that rate 2. For the sake of 3. come to 4. by hand 5. vice versa （八）、1. 年轻人有时会抱怨无法和父母沟通。(communicate with) Answer: Young people sometimes complain of being unable to communicate with their parents. 2. 能在中国云南的一个苗家村落住下来一直是玛丽长久以来的梦想，现在她终于梦想成真了。(to take up residence) Answer: It has been Mary’s long cherished dream to take up residence in a Miao village in Yunnan, China. Now her dream has finally come true. 3. 家养的动物习惯于依赖人，因此很难能在野外活下来。(survive) Answer: Domestic animals are used to depending on humans, so it is difficult for them to survive in the wild. 4. 他突然有种恐惧感，觉得自己会因为经济不景气而被公司裁员。(overtake) Answer: He was suddenly overtaken by a fear that he would be laid off by the company because of depression. 5. 我估计公交路线上堵车了，因为我都等了30分钟也没见一辆车开过去。(figure) Answer: I figure that there is a traffic jam on the route of the bus, for I’ve waited for 30 minutes without seeing one passing by. （九）、Answer: In spite of trade frictions, the US cannot do without China or vice versa. Answer: If you find yourself working without efficiency, you’d better relax for a while. Answer: The speaker glanced around the audience this way and that before he started speaking. Answer: My father’s opinion on my job change sounded to me like a warning. Answer: The labor union agreed to give up a pay increase for the sake of more job opportunities. （十）、Answer: The sandstorm is a horrible phenomenon. The clouds of dust are so heavy that
Unit1 1.semiliterate:a.semi-educated;having only an elementary level of reading and writing ability半文盲的，有初等文化的semi-:half or partly 2.dropout:n.a personwho leavesschool before finishing a course（尤指中学的）退学学生 3.do-gooder:n.sb.who does things that they think will help other people,although the other people might not find their actions helpful一个总是试着帮助别人的人（通常是贬义）；不实际的社会改革家（指幼稚的理想主义者，支持善心或博爱的事件的改革者） 4.impediment:n.obstacles,barriers妨碍，阻止，阻碍，阻挡 5.trump card:anything decisiveheld in reservefor useat a critical time王牌 6.charmer:n.a personwho hasgood qualities that make you like him/her讨人喜欢的人，有魅力的人，有迷惑力的人（尤指迷人的女人） 7.get by:to be able to deal with a situation with difficulty,usually by having just enoughof sth.you need,suchas money过得去 8.settledown:to becomequiet and calm（使）安静下来；平息 9.flunk:v.to fail an exam or courseof study不及格 https://www.doczj.com/doc/db11836455.html,posure:n.calmnessandcontrol平静；镇静；沉着 11.parade:n.a seriesof peopleor things that appearone after the other 12.at stake:to be won or lost;risked受到威胁，面临危险 13.sail:v.to move quickly and effortlessly投入 14.testimony:n.spokenor written statementthat sth.is true证词，证明 15.conspiracy:n.act of joint planning of a crime阴谋，共谋 16.doom:v.to makesb.or sth.certain to do or experiencesth.unpleasant注定 17.follow through:to continue a stroke,motion,plan,or reasoningthrough to the end 将动作、计划等进行到底 18.I flunked my secondyearexamsandwas lucky not to be thrown out of college. 19.The managementdid not seemto consideroffice safetyto be a priority. 20.Thousandsof lives will be at stake if emergency aid does not arrive in the city soon. 21.I think therewas a conspiracy to keep me out of the committee. Unit2 1.propose:v.[to sb.]to aska personto marry one提亲；求婚 2.knockoff:n.a copy or reproduction of a designetc.esp.one madeillegally假货；赝品 3.Windex:v.to clean with a kind of detergentby the brand of Windex用Windex牌清洁剂清洗 4.takethe plunge:to take a decisive first step,commit oneself irrevocably to a course of action冒险尝试 5.bridesmaid:n.a girl or unmarried woman attending the bride at a wedding女傧相；伴娘 6.maid of honor:a principal bridesmaid女傧相
个人与企业的关系 企业发展离不开各个方面的发展，但就个人发展与企业发展的关系而言，从某种角度来说，可以理解为鱼和水的关系。鱼儿离不开水，也就是说个人的发展离不开企业这个大舞台；水深则鱼悦，表明企业发展良好，就会给个人发展提供相对较好的条件，个人从中受益匪浅！在现实社会中，我们每个人要给自己定好位，为企业的良好发展贡献自己的一份力量。 只有在集体中,个人才能获得全面发展其才能的手段,也就是说,只有在集体发展的前提下，个人才有可能发展 集体的氛围和风格在无形中会影响到每一个人，举一个最简单的例子，就是大学的文化和历史沉淀，人们常说在北大呆两年，什么也不干，熏也给你熏出点北大的气质来；还有“近朱者赤，近墨者黑”，其实都是这个道理。 其实，说的这些很多人都是认同的，并且在意识里是存在的，但是往往却有这样的一个误区，总希望自己成长在一个优秀的集体中，却从没有考虑怎样去营造一个优秀的集体，怎样去领导一个集体，或者当面对一个集体的困难时，不愿意去承担责任。当然，这中间可能有自身性格、能力、取向等各方面的原因，但是我想从根本上来讲还是没有把集体的利益真正融入到个人发展当中去，还是没有正确的团队观念。 那什么是正确的团队观念呢？每个人的看法都可能不同，但我想至少两个方面，一是主观欲望上渴望在一个优秀的集体中，并愿意配合集体的各项活动，愿意为集体的成长进步尽一个成员的义务，希望集体的发展促进自己的成长与进步；二是愿意承担集体的责任，以集体的发展为己任，以集体成员的成长进步为己任，尽自己最大的努力促进集体的发展。其实这两个方面说简单一点就是成员意识和主人翁意识，但二者却有着本质的区别，应该是两个层次上的。 我想大学时代，最重要的集体就是班级，大多数同学对班级都是有感情的，但往往都是停留在第一层次上，都渴望自己的班级团结、向上、相亲相爱，但很少有同学主动的去营造这种氛围。其实班长和普通同学在班级建设过程中的角色定位的差别就在这里。很多同学不愿意做班长也是因为这个，不愿意去承受这种压力。 但我想，任何一个团队，任何一个集体中，如果成员只是具有第一个层次的团队观念的话，这个团队不会取得多么大的成功，这个团队的成员也不会有什么大的发展；而如果每一个人都把自己当着团队真正的主人，都在谋划团队的发展，我想团队就会越来越好，成员同样就会越来越好，这就会形成一个良性的循环。 集体的发展是每一成员都应该思考的事情，每一个人都会从集体中获得东西，有形的或无形，如果你