我们知道，因为科技的关系，这个世界的先进程度早已经超出了你我的想象。
所以苹果能跟梨杂交， 乌鸦能跟鸽子撒娇，好好的守护BOSS都能变成TVB的腔调，
那么好有什么不能改变？
一个没有图像没有画面的音频，假如被剪辑合成的乱七八糟之后，
还会呈现什么样的面貌？
下面为你揭晓！
=================================================================
Owing to science and technology, the world is being developed more rapidly than we imagine, many impossible dreams have been realized, while an increasing number of evil doings appear and are performed more easily. （此处只翻译大意，并未逐字翻译）
What one audio without pictures or images will be turned into, if intensively edited and composed?
We will show you as follows:

1.【模拟还原现场，疑点重重】
@STILL_in_LOVE_KJJ：[还原现场]今晚刚做好的，让同寝室的妹子友情出演了⋯“假设私生被打，撕扯中录音”在这样的假设下真实模拟了所谓“现场”，被打的人头痛死了⋯录音设备是SONY+ICD-UX512F+2G+大陆行货价格1080RMB，我相信私生的录音设备不会比这个更好了吧？（大家不要笑⋯）
==========================================================================
1.【Simulation and Reconstruction of the Scene, Many Doubts Are Found】
@STILL_in_LOVE_KJJ: [Scene Reconstruction] I ask one of my roommate to help me together record this audio. Supposing that the saseng fan is hit while recording the sound, we simulate the situation. The recorder is SONY+ICD-UX512F+2G at the price of 1080RMB, which is certainly no worse than that of saseng fan. Please listen.
（未逐字翻译，大意翻译）



加一楼音频中的文字翻译（为逐字翻译，大意翻译）
Words on screen:
The recording place: student dormitory without noises;
The situation: A grabs B’s left arm and heavily hit B’s head. B screams out of pains. This is recorded for testing noises in this situation.
Equipment: SONY+ICD-UX512F+2G at the price of 1080RMB
Scene 1: A third party records the sounds closely. We can hear obvious noise of bumping.
Scene 2: The recorder is kept in pocket. We can hear obvious noise of attrition.
Doubt 1: After recording we can see that girls normally scream in a high-pitched sound when feeling pain. But the saseng fan’s sound in JJ’s audio is investigated as no reflecting pains or fears. So how can one girl react that calmly when beinghit?
Doubt 2: The possibility of keeping the recorder in hand when recording JJ can be estimated as zero, so there are two scenarios: a. A third party is on the site recording within one meter distance from JJ. On that condition the third party must be influenced when the saseng fan is hit and there cannot be without any noise of bumping. In reality the third party is impossible not to produce any sound. b. If the recorder is kept in the saseng fan’s pocket, during the hit, it will certainly records the noise of attrition. This noise cannot be removed by noise reduction methods.
Conclusion: The audio has many doubts against truthfulness. By one simple recording we can see some points. Our hearts and endeavors to protect JJ from ill imputation are unanimous.

2【声纹分析实时拍摄，证明音频为合成，且技术粗糙】
F:mjjeje小淫
看视频。该视频是软件对流出的音频进行声纹分析的过程进行的实时拍摄。当有人声出现时，右边框内就会有红色声纹出现该音频由三部分剪辑合成，当第一段和第二段（类似在中声音部分）衔接时，右边的红色声纹宽度区别很大，第二段的宽度明显比第一段的要宽；同理，第二段的声纹宽度也比第三段的要宽。该种迹象即为剪辑合成音频。另外，要在音频中插入其他声音，只需复制粘贴你所需要的音频元素就OK了。
简单一句话，视频证明了是合成的，而且对方的合成技术也太粗糙了点
=======================================================================
2.Record Shooting of Voiceprint, Proving the Audio is Composed roughly
F:mjjeje小淫
Please look at the short shooting. This video records the analyzing process of the voiceprint in the audio. When human voice appears, the red voiceprint will appear. This audio is composed of three parts. At the connection point of Part 1 and Part 2 (appearing the voice similar to JJ), the red voiceprint on the right appears in a large width, much wider than that in Part 1; likewise, the red voiceprint in Part 2 is wider than that in Part 3. This phenomenon shows that the audio is edited and composite. What’s more, if you want to insert other sounds in the audio, it is just a work of copying and pasting the sounds you need.
To sum up in a simple sentence, this video proves that the audio is composite and the skill is rather rough.



3【声纹分析也可以证明音频经过多次剪切合成，场景的变换十分之诡异】
3.The Analysis of Voiceprint Also Proves the Audio is Edited and Composed for Many Times; The Recording Background Weirdly Changes Many Times
FROM:忙碌中的小喇叭
【关于第一段音频】图片有点小，仔细瞅：首先，1'30是一个很明显的分界线，1：28:7到1:29:8之间是空白，说明这两段是被拼在一起的。第二点：1‘30之后的波长基本没有起伏，而1’30之前波长起伏很大，如背景杂音为真的话，波长是不会有大的变动的
FROM:忙碌中的小喇叭
[About Part 1 of the Audio]: Please see the picture below: Firstly, 1’30 is an obvious dividing line, from 1:28:7 to 1:29:8 it is blank, illustrating this two sessions are composed together. Secondly, wavelength after 1’30 nearly stays same, while before 1’30 undulates obviously. If the background noises are recorded at the same time, the wavelength will not change a lot.

大家按照上面给出的时间点的提示来看，可以看到波纹有明显的断裂拼接的痕迹。前后的声纹波形相差也很大。如果是在同一个场景下发生的话，是不会出现这样明显的断裂的。录音这玩意儿相信大家都试过，在一个开放的环境（比如这个音频当中假设的是在路边）无论怎样都会有电流声听起来背景音都是有些嘈杂的，不可能如下图标出的点上这般出现静默
Please see the time points in the picture, traces of disconnection and splice of the voiceprint are very clear. The wave shapes of the voiceprint in two parts divided at 1’30 are much different. Everyone knows that if in an open environment (e.g. this audio is supposed to be recorded on road), however you record there exists a sound of electric current, so the background sound is a bit noisy. Such silence cannot happen as reflected on the point marked.


FROM:忙碌中的小喇叭
【关于第一段音频】再来看24’0-26’5这段，前面的波长多和谐啊，波长较厚那段是车子声音最大的部分，26‘2的时候还听见车子呼啸的声音，但是26‘3-26’4之间那一竖线多诡异啊，前面的声音突然消失，只听见“哒”地一声，太尼玛神奇了，车子的声音突然消失（或减弱）了0'2，26’5之后又恢复过来了
FROM:忙碌中的小喇叭
[About Part 1 of the Audio]Again let’s see the session from 24’0-26’5, wave length of this session is natural, the wider part represents the loudest car horn. At 26’2 the sound of cars can still be heard, but from 26’3 to 26’4 we can see one weird vertical line in the picture, and the car sound before suddenly disappears, and only occurs a sound “Da”. It is weirder that at 26’6 the car sound recovers after sudden disappearance(or reduction) for 0’2.

FROM:忙碌中的小喇叭
【还是24’0-26’5这段】再次放大，只看26‘18-26’35这段，声音的波长显示是波谷-波峰-波谷-波峰这样循环的，请仔细看图中白色那段，再对比前面的波峰波谷，明显白色部分的波峰木有了！太监了啊有木有！紧接着后面就是26‘33-26’34那个很诡异的竖线也就是我们听到的“哒”，很明显动了手脚
[Again the Session from 24’0 to 26’5] We enlarge this picture and focus on part from 26’18 to 26’35, where the wave length appears in a cycle of trough-peak-trough. Please carefully look at the part in white and compare it with the part left, it is clear that the peak of the white part disappears! Then from 26’33 to 26’34 appears the weird vertical line, where we can hear the sound “Da”. Apparently this session is processed.



4【波形分析音频为剪切拼接而成，且背景音很多地方都没剪切干净，漏洞百出】
Based on Wave Shape Analysis the Audio is Proved Edited and Composite; and the Background Sound Has Many Unclear Cuts

FROM死人头小组
We know that if you make a sound recording without changes in place, it has one background sound. Please look at the wave shape analysis below. At about 26’the background sound turns louder. Maybe it can be explained as there is a car passing by at that moment, but how does a normally running car take several minutes to go some meters? Also aiming at the last part about hitting, which is the fiercest, its wave shape is much different. From this we can see obvious tracing of editing and adding in this part. (为图片中文字部分翻译；梳理中文后再中翻英）

5【咨询音频方面的专业人员，得出音频为合成的结论】
Consulting Professionals on Audios and Concluding this Audio is Composite
F 微博 闯祸不闯祸不是闯祸
上面蓝字的是微薄的某位亲以两分多的那个音频找淘宝买家得到的，我们注意到这位提到了我们从来没有注意的点：那个女人的声音很假很清晰而且无痛感无惧怕！！！想想看处于这种情况下的人怎么可能会无痛感无惧怕呢？
F: @ 闯祸不闯祸不是闯祸The screenshot below shows one fan consulting a professional one part of the audio lasting about 2 minutes, and is reminded of one point: The saseng fan’s groan sounds fake and reflects no feeling of pain or fear!! How can one girl react in this way after being hit?
F:_Sssu_ 注（蓝色图片部分）
F:_Sssu_ (Note: Words in blue)

Translation of the blue words in the picture (图片中蓝字翻译）
I doubt the girl’s groan. It is so clear, without any noises and showing no feeling of pain. So here is the conclusion:
1. The audio is edited at 23rd and 77th seconds;
2. The girl’s groan sounds fake, it is clear, without any noise and showing no feeling of pain or fear.
那么真正的挨打的人的反应会是什么样的呢？请参见3楼的模拟现场
Comparing with a real reaction when someone is hit, please see evidence 1—The Reconstruction of the Audio.


【对比分析相似度不高，疑为后期添加合成所致】
Rate of Similarity to JJ’s Voice is Not High, Doubting This is Leaded by Later Adding and Editing
以coo片段金在中声音为基准，音频1的平均相似度为0.76，音频2的平均相似度为0.6825，音频3的平均相似度为0.75，音频4的平均相似度为0.545（用金在中其他音频测试，一样与coo对比，平均相似度为1.00）
Taking JJ’s voice in COO as the benchmark, the average similarity rate of Part 1 is 0.76, Part 2 is 0.6825, Part 3 is 0.75, and Part 4 is 0.545 (Taking other audios of JJ to compare with that in COO, the average similarity rate is 1.00).



F:楠楠-魊_我不是思密达:
刚才去找了一下人分析了一下第一段音频，人家说，中间有几段是完全无声音的，也就是说即使不能证明不是同一段时间录的，也可以证明不是连续录的或者进行过调整……人家说话保守，大家完全可以自行想象
F:楠楠-魊_我不是思密达
Asking professionals to analyze the audio and finding that some sessions are without any sound, which proves at least that this audio is not recorded continuously or rearranged.(未逐字翻译，大意翻译）
音频1里,首先看一下8到11秒这一段~图有点小,但是可以看到有一条完全没有波动的直线,就是说这段时间是完全没有声音的!说明什么啊，拼接剪辑神马的嗷嗷~~~
Part 1: Firstly look at the session from 8’ to 11’, here appears a straight line without waves, illustrating that there is no sound during this period. Here trace of editing is clear.

同理再看16到17秒，还是一道跟死了人的心电图一般的直线，完全没有声音是怎么样形成的啊~~~
Likewise please look at the session from 16’ to 17’, here the straight line appears again and illustrates that there is no sound again.

最后我们再来看一下1分28秒到1分30秒这里，咱们当示范剖析一下啊。、可以看到，音频中间有一段明显直线，没有任何声音波动，而且是发生在音频的中间段，就是很明显的拼接了吧~~~
Again we find from the session 1’28 to 1’30, in the middle appears an obvious straight line, here the trace of editing is so clear.

@楠楠-魊_我不是思密达：我妹那边不给力，所以自己只能逼自己化身技术帝！昨天就有人指出音频1和3里两个挨打的声音a是何其相似，于是我截了两段的音轨，一样啊，完全一样啊！请看音频1的40秒之后和音频3大约1分55秒开始的对比!你们污蔑吧,我什么都不信!!!
Pointing out the groaning sounds in Part 1 and Part 3 are pretty similar, so I compare the audio tracks of the sounds. They are totally the same!! Please look at the comparison below about groans at 1’40 in Part 1 and 1’55 in Part 3.



现在整理出来的证据暂时先发到这里了，其实熟悉在中的人都明白，他是个很关心饭的人，也确实会对饭说教，说教的内容也没问题是人之常情，但是真真假假的音频混起来，就是为了让大家误认了打人那段音频也是在中的
这音频要达到黑他的目的必然是半真半假、拿他说过的话和打人的声音合成、造成打人的假象。以上种种，都明确证明了这个音频是被人动过手脚的掐头去尾的添加了“特效”的合成音频。那么这样一个合唱音频又有什么样的价值让人去相信呢？
前面的证据已经证明了音频是恶意合成的，是故意陷害在中的。这就是一场阴谋一场陷害！不管陷害金在中的到底是谁，我们只要知道，在中是无辜的。在中对饭有多好，这么多年了我们还不知道吗？只要对在中有爱的饭，对伤害在中的事情都是不能容忍的，所以我们一起站起来，一定要尽力守护好在中！
感谢参与到这次澄清中来的我们最可爱的花粉以及真心爱着在中的为了制作视频图片努力的团饭们，大家的这次努力都会被铭记！
后续还有重重疑点，都会发上来与大家共同讨论如果有新的证据也会在楼里更新的，也希望大家积极发现新的疑点与证据，还我们在中的清白！
=======================================================================

(发帖有问题，文字版请PM）

==================================================================
以上为现有楼层证据翻译以及略微补充；
希望能够多多讨论并且尽可能完善；然后发挥力量扩散！
让金在中安心准备演唱会，澄清的事情就交给我们吧！与黑黑们的时间赛跑，走起


真的是辛苦楼主了