情形於聲通訊#2 野外錄音手記

本期封面推薦是 @ Digimonk《野外錄音手記》,乾貨滿滿,建議熟讀。【目錄】活動、作品;設備;敎程、技術。

Activity & Work

LISTENING TO RIVERS 一場網上會談 Wednesday, September 9, 2020 3:00 PM 4:30 PM This event takes online via zoom

@ Félix Blume, 2020-08-09


@ George Vlad 2020-08-27

Mantled guereza or Colobus monkeys in the Harenna forest, Ethiopia. I was woken up by their terrifying calls on my first morning in the rainforest. https://www.youtube.com/watch?v=rtVT5KKViN4&feature=youtu.be

@ The Recordist 2020-08-20

A couple of clips from The AK: Kalashnikov Rifles SFX Library video… Sound is a mix of only 2 microphones, a close up MKH416 and a MKH8040 ORTF set. No other processing, just some limiting to amp it up a bit.



4060, 8060, H4N,

ANURAG 4060 非常自然,低頻非常棒;對風噪不敏感。

A railfan from India here. I have experience with PCM-M10, H4NPRO, MKH8060, NT4 and 4060. I feel the 4060s worn around ears beat all others in naturalness. If I were starting today, I would have bought just a mixpre-3 and a stereo 4060 kit, passing all others. Trains are loud and noisy, so mic self noise is not a big issue when recording close passbys. But transient response and bass extension are, and the dpas excel there. 4060s are not much sensitive to wind, so I can get usable recordings of not so fast trains with zero wind protection on calm days(by standing like a statue). 4060s can also handle high spl with ease, after I dial down gain to about 15 on Mixpre-3.

A comparison of the above kits from an amateur’s perspective:

PCMM10: Good bass extension (comparable to 4060s), good wind resistance with deadkitten, but bass is boomy and loose at times, overwhelming other subtle hf details. Transient response is so so, it cannot handle high spl at all (some digital distortion occurs when gain is reduced to avoid clipping). Abysmal stereo separation due to two closely placed omni mics. But the most portable and practical setup for quick recording.

H4nPro/NT4: They are very similar in sound. Very weak bass extension, yet insanely senstitive to wind and handling noise. NT4 has much lower noise, but also slightly darker. Both of them sound harsh when recording high spl train horns, but while NT4 is usable, H4n is not. I found H4nPro stereo separation to be much better, even at 90 degree mode. At 120 degree mode, centre suffers (i.e the cardioids sound dull when 60 degree off axis response).

8060: Sounds dull if off axis, has a very narrow angle of sweet spot. On axis sound is slightly honky (hyped midrange).However, it has low noise, good transient response and reasonable bass, good hf rejection from back and sides that become lifesaver in noisy areas and recording targeted sounds (e.g. just the loco, not the carriages).

@ Digimonk 的微博

每次看到 Sonosax 都让我双眼放绿光。前段时间体验 M2D2,Sonosax 底噪特别低,在 gain 非常大的时候也听不到多少噪声。SD 完全做不到,833 的 gain = 35(12 点位置)时候,fader 开到 +10 就极限了,再大一点就可以听到底噪了。如果环境声在 35 以下的话,这个真的很要命,833 录到的环境峰值不会超过 -30,在后期正常监听音量里几乎听不到声音。92kHz 录音后期提升 12dB 之后可能还凑合,如果 192 录音就太麻烦了。

一套 Schoeps DMS 在 Cinela Pia 3 罩子里,一台 833 在 Orca 272 背包里,LOM Usi Pro 话筒一对,4 块 Hi-Q 98 电池,Cinela 长毛防风罩,两套 Mogami 线的 DMS 话筒线,一副 HD25 II 耳机,一副 Shure 535 备用耳塞。小米插座板、Pelican 防水盒里都是备用件。防雨罩、头灯、电池充电座、3M 胶带、Suri 三脚架。另外随身带一台备用 SD Mix Pre6 录音机。

Cinela 风罩里是一套 Schoeps DMS 主话筒,横架上是一对 DPA4061 录备份。全部进 833。一块 Swit98 电池,两块 F750 备用电池。全部装进 Orca272 背包里。可能是目前最紧凑的高品质录音方案了 [可爱]

2020-07-20 我的筆記:

Schoeps 阻抗大,帶來錄音機底噪。744、833 推力大,但 Mixpre 底噪大。Schoeps 聲音不好聽,所聽卽所得,抗震不好。cinela 不用毛也可以抗 7 級風,而且抗震。推薦新手用 lom usi pro。一個小毛衣,十米風沒問題。話筒超過自己動態範圍會破音



I would go for the Seven - the LB was not designed as a music machine, it was designed as a reporter’s recorder. which is why is has the built-in editing functions (but no good for music as it’s for the compress files and low quality wav files) and had the ability so send recordings back to base.

The Seven has higher quality pre-amps and, for a 2-channel machine at that budget level, I would go for the Seven.

Sound Devices

I do prefer the pre’s in the 744T and 664 to the 552, though

788T 的話放和 RME Micstasy 一樣:

SD writes:" Additionally, these new preamps (788) are fully digitally controlled." Then you know almost for sure, unless they have created their own design, they use the Analog Devices PGA2500 chip. Same is e.g. in the Micstasy of RME. SD mentions the 702, 722 and 744 use discrete transistor pairs. I prefer discrete transistors, lower noise with lower gains.

Quality Preamps/AD of Sound Devices 644 vs 788 744T 比 788T 聲音好:

@ RPSharman: I’d use a 744T - best pre, best limiters, best all around.

@Constantin: I agree. If you don’t need any if the other features of the 788T, the 744T is best. And it can be bought at a fairly low price used nowadays

探索 833/744T/MixPre6 录音机之间的差异 | 833 国内首例专业对比测评

MP6 拾取了不仅更多的高频,也拾取了更多的低频。

Mixpre 有明顯的染色,聽起來比較好聽:

So we have these three Sound Devices recorders: 1) 788T, 2) 744T, and 3) MixPre3. Decided to do a studio shootout @ 48K and 24 bit. The sound source? A Cordoba Flamenca Negra GK pro played by yours truly, recorded through a Mackie 1402 VLZ3 with NO EQ and FX provided by the lovely Eventide H9. The performance? Combination of finger picking and strumming with alternating dynamic range to test transient response.

The results? Both of us ranked, from best to worst: 2-1-3. Why? Basically, we both felt that the 744T brilliantly picked up the transient responses of the Negra, and the 788T was a close second. 744T won again on linearity; again, with the 788T being a close second… so what about the MixPre3?

Well… my partner said “The MixPre3 would be a great recorder for a beginner, because it makes the performance sound better than it actually is.” Not a very flattering comment when judging a recorder

Basically, this recorder had noticeable “smearing” that was particularly present in the low mids. It also had this effect of “smoothing over” finger noise and other subtle transients that, for me at least, give a performance its mojo.

Were the differences huge? Let’s put it this way; unless you were a trained audio pro, probably not. Would the MixPre3 suffice? Probably would do just fine for dialog recording–but for my money, I wouldn’t use this recorder for professional music recording.


M32, R4,62r 的話放完全一樣:

I heard from Sonosax that the preamp and AD converter is exactly the same as the SX-R4.

I tried the Sonosax M32 mixer, a three channels mixer having the exact same microphone preamps as the 62r.

新的錄音機和老的 mixer 話放完全不一樣,但都很好:

I have only used the older model of the Sonosax and only used it for one recording that I did in Pittsburgh, PA and all I can say is WOW! Great sounding preamps, velvet background and tons of headroom. If I had the money I would buy one in a heartbeat.


I did a comparison a while back with the SQN IVe, Sonosax SX-M32, Sound Devices 552 and Fostex FM3. IMO preamps for quiet/cleanness (tested with a 416):

Sonosax——SQN——Fostex (although v close to SQN!)S——ound Devices

windshield & clip

BubbleBee Industries 防風滑套

DIY windshield for LDC mid-side pair


Rycote 的支架,用於小型錄音筆。




广州影视同期录音师出租!接活 自带设备!清单如下:

  • sound devices 6 录音机
  • SONY 监听耳机两对
  • SENNHEISER/森海塞尔 MKH8060
  • SENNHEISER/森海塞尔 MKH416
  • SENNHEISER/森海塞尔 MKE 600
  • 英国灵巧 Rycote 超级防风套装
  • Rode REPORTER *动态面试麦克风
  • RODE 防风毛衣
  • 小蜜蜂 2 套
  • 手雷一个
  • 音频发射一拖三
  • 可直接入机内录,支持双机
  • 导演监听 red,mini,fs7,bmpcc4k,a7 等
  • 国产碳素挑杆 3 米、5 米各一支
  • 录音设备手拖箱(可放下上面所有设备与线材电池等)除挑杆外

设备,500 一天。录音师加设备,1000 一天。录音师加设备加杆爷,1500 一天,可自由搭配。价格私我,长租优惠。外地报销路费。


大工程,采风录制时间大概 1-2 个月,需要能接录音的公司或者团队,自带设备 具体要求如下:

  1. 8 通道录音声卡一个
  2. 主麦克风 (电子管) 两只 (并搭配各类野外录音防风罩)
  3. 立体声小振膜多指向性麦克风两对 (四只,并搭配各类野外录音防风罩)
  4. 动圈或铝带式中低频麦克风 - 只
  5. 与七只麦克风搭配的话筒放大器 (七通道)
  6. 监听耳机十副&耳放
  7. 便携移动型声学处理模块
  8. 高端便携式小型数码录音机两个
  9. 1200W 锂电池供电系统 (可汽车逆变充电或太阳能充
  10. 重型麦克风支架五只
  11. 苹果电脑,以及外置硬盘,备份硬盘若干
  12. 小型监听音箱一对户外三防器材保护箱若干

(二) 采风录制标准 1.24K/192hz、7.1 声道、High Resolution 数码音频 2 录音格式:WAV24bit,ST,48(国际出版标准)


Nevaton MC59

Nevaton MC59 self-noise is only 5dB!!!

The surprise in this comparison is the Nevaton MC59O. I finally found a omni mic which has lower noise floor than MKH20, without loosing the finest details, plus with extra sound clarity above 10Khz.

ultrasound microphone

Condenser ultrasound microphone Frequency range : 2 kHz – 200 kHz!!!太神奇了

Parabolic Microphone

Wildtronics Professional Mono-Stereo Parabolic Microphone Equivalent Self-noise Stereo: 5dBA

CUW-180 Surround System

CUW-180 Surround System,神奇的麥克風,神奇的錄音方式。淘寶上根本沒得賣。

The purpose of the line microphone CS-1 is to secure the independence of the center channel and to minimize the overlap of L and R. The feature of the CUW-180 surround system is to keep the natural continuousness of five channels without dead points, and the accurate localization of sound images.

Clippy XLR EM172

4 Matched Clippy XLR EM172 Microphones

4 箇全指向微型麥克風,用來錄環繞聲,非常便宜,共 £ 229.92。大概可以當成丐版 DPA

contact mic

JrF c-series contact mics 接觸式麥克風,不貴。

Geofón Geofón is a sensitive geophone adjusted for field recording purposes. Originally designed for seismic measurements, it can be used with regular field recording equipment to capture very faint vibrations in various materials and even soil. 用來接收振動的麥克風。

Tutorial & Technique


2019-08-24 ● 建議熟讀

丛林、风与季节:合适的时间,合适的地点真正的荒野丛林里大部分时候是非常安静的,差不多可以和录音棚一样安静,甚至可以低到 29-35dB,很多录音棚都不一定能达到。可能打破这种静谧的基本只有鸟鸣和风。鸟鸣主要是看时段,清晨 5-7 点、下午 4-6 点,通常是鸟类进食时间,那时候声音会比较多。但那种声景只是自然界的一面。在静谧的树林里,偶尔出现的几声鸟鸣,因为丛林特有的混响,也会变得非常有意思。虽说丛林的静谧,其实也是有底噪的,主要来自风。风推动了树干的摇晃、树叶的摩擦和抖动,风也在林子里推动气压差而产生空气和植物的摩擦。丛林的密度越高,那么就需要更强、持续时间更长的风才能驱动声音产生。夏季,尤其是白天,大部分地区的丛林都不会有什么风,3~4 米/秒的风,基本上不会产生明显的声音,即使在山地也是如此。但是夏季的夜晚,冷热空气的交替会产生比较强的风,所以要做好在林子里过夜的准备。人工林的夜晚虫鸣相对来说比较少的,尤其是北方地区。但是在南方,比如说杭州,夜晚的丛林比白天热闹多了。像新疆乌苏那里,尽管禁猎多年,但是鸟类活动还是要到远离人类聚居的深处才会多一些。大型鸟类、尤其是肉食性鸟类,其实基本不怎么叫唤,你能听到就算运气爆棚。松林,比如说新疆的雪松林,会需要更强的风才能引起明显的声音,例如大家熟悉的“松涛”。如果风弱的话,首先推动的树梢,所以话筒要向上一点。那种风主要是入夜后冷热空气对流产生,所以不是很稳定,听起来也会像海浪一样一波一波的。所以出远门录音前,最好连续观察当地的气候、以及同一季节在过去两三年里的局部气候状态,这样才能相对准确地预估可能录到的素材。秋、冬季节可能是最适合录丛林声音的季节,强劲的北风、西北风会让密林产生很强的轰鸣声。尽管如此,到了当地也依然要花很多时间去寻找合适的录音地点,并不是每个地点都会有你需要的声音。这里有一组简单的经验数值;

1 m/s:小到你的脸都可能感受不到的小风,很柔和,连草都不会动。安静也是一种极其难得声音。扒下防风毛套,开大增益试试。但是建议不要把话筒裸露在外面,放在笼子里可以有效抵御偶尔传过的瞬时强气流。录这种声音,是很考验话放和话筒性能的。这种声音往往会很像话筒和话放自身的底噪,也是需要训练自己判别这类声音的不同。

3 m/s: 依然很柔弱,脸部偶尔能感受到一些气流。草叶也不会动,最多偶尔摇晃一下。但如果开大话筒的增益,你能听一些底噪的。尤其是在密林里,还是要会产生一个持续的低频底噪。如果话筒合适,这个声音也是很漂亮的,有明显的大空间感。(越是低速的风,它的稳定性越差,分布也越不均匀。)

6 m/s:还是很弱的风,但是脸部可以轻易感受到了,也足以催动草叶、树叶动起来。在丛林的外围、尤其是树梢部分,是可以产生明显的线声源的。这时候录音,一定要带上长毛套的。阵风可以轻易穿透Rycote风罩、直接把话筒吹爆,哪怕把LPF开到70Hz都不管用。有些阵风的面积很小,你站立的位置或许根本感受不到,但是它可能会穿过话筒的位置,这也是弱风天气的明显特点。个人建议,如果在山地,天气预报说风速3m/s的时候,就要带上毛套。如果有短毛套最好,如果没有,用长毛套也问题不大。(我也只用长毛套,箱子里少一件东西都是好的。)这时候还需要考虑另一个技术处理:LPF。无论是话筒还是录音机上都建议考虑一下是不是要打开。个人建议还是打开会比较安全,可以开到42Hz或者70Hz。除此以外,最好找一个避风的地方放置话筒,不要直接面对风。尤其是在山地,会产生强烈的瞬时阵风,长毛套都不一定能扛住。

8 m/s:天气预报显示这个风速的时候,其实是比较强了,足以让细树枝、树梢小幅度摇晃起来。松林,是最不容易被风摇晃的,这时候也会有明显的反应,主要来自树梢和风在树林里的气压变化。长毛套戴起来。如果有强风罩的话,可以优先考虑用单层强风罩,以对应可能产生强阵风,千万不要高估长毛套的抗风能力,它的能力很有限。控制增益的强度,不要推高(偶尔的强阵风会导致树干、树枝产生碰撞、扭曲,这样的声音很漂亮,但是峰值可能会很高,所以增益要留出至少16-20dB的空间)。找一些合适的、背风的录音点,多录一些,每个地点不要低于10分钟录音。最好踩点的时候就留意好这样的地点。圈内说,同一个录音区域,不同的录音师总是会发现自己喜欢的甜点。

10 m/s: 绝大部分松林、丛林都会开始产生类似海浪的咆哮声。风速、风向、地形、植被分布都会影响这种声音的起伏频率和强度。一般来说单层强风罩是可以扛住 10m/s 的风,录音时候一般建议话筒是背风方向的、甚至是有遮挡的位置。但有些时候如果只能让话筒迎风站那里的话,还是在强风罩上套一个长毛套会比较安全,这要看具体地点的阵风情况。另外需要考虑的是话筒架的稳定性了,用石块啥的压住架脚。

13 m/s:这个风速意味着接近风暴的强风开始了,一定要找背风的位置和角度摆放话筒。强风罩外面加一层长毛套,应该是足矣了。除非作孽遇到龙卷风啥的,不然完全可以扛住的。另外需要关心的问题是话筒架和你自身的安全。其实,风越大,越建议你往林子深处跑,那里相对更加安全,也更容易捕捉有深度和宽度的丛林声音。如果身在北方,这样的强风一般来说相对比较安全的,但是如果在南方、尤其是夏季,强烈建议谨慎出行!强风意味着雨水,降雨会带来泥石流、洪水。


​ • 当你脸上感受到持续的风,哪怕是很柔和,那么最好把毛套起来 ​ • 地面上的草叶大面积的动起来,即使是一阵一阵的,最好把毛套也穿上。 ​ • 看到地面上的灌木植被一直在被风吹着晃,毛套可能就没啥用了,改用强风罩会比较安全 ​ • 如果脸上感受到强风,即使是阵风,建议强风罩再加毛套一起穿上

风罩,无论是毛套还是强风罩,都是会损失高频的,尤其是像 Neumann 145、Earthworks TC-20 之类的高灵敏度话筒,防风套对高频的衰减会很明显。这也是我们需要在不同情况下使用不同的防风策略的主因。但这些 16Khz 以上的高频衰减,也不见得是不好,其实大部分时候在影视和游戏里根本不会需要这样的频段,只有一些特定剧情和场景才会强调这部分高频,那时候可以在后期动用 EQ 来提升和修整这部分被衰减的高频(建议使用动态 EQ 来处理)。

左起:Rycote 强风罩、长毛套,AEWS10 悬架防风罩

左起:Cinela 内衬防风面罩、长毛套、Piani-3 DMS 悬架防风罩、防雨滴海绵罩

对于 Sony D100 之类的手持录音机,包装里附赠的毛套对这样的录音来说就是形同虚设的,可能连最小的阵风都扛不住。Rycote 有为手持录音机设计的悬架和防风套,还是非常有效的。但即使如此也扛不住 8 m/s或者更强的风。如果条件有限,那么唯一方法是找到尽可能避风的地方录音。不要相信广告里迎风而立录音的样子,那除了装X还是装X。


​ • 再次检查装备是否安全妥善的固定好了、可以方便运动了。 ​ • 再次检查衣物、鞋子、备用器材是否完整,是否有哪些可以不必携带 ​ • 再次确认电池、线材完好 ​ • 预估徒步时间,估算需要携带的水量,准备一些巧克力补充糖份和热量。


在高海拔地区,例如 1800 米以上的乌苏,夏天的树林、草原和草甸上安静的几乎可以听到自己的心跳。但依然会有瞬时阵风,这样的阵风分布非常不均匀,也因此会产生不一样的声景。可以把话筒放低,对准那些略高一些的茅草从,当风吹过的时候它们会产生很漂亮的摩擦声。仔细听耳机里的反馈,调整增益,寻找合适的距离,就可以获得多层次景深的声景。因为阵风的分布不均匀,但又要确保话筒里收到的声音在立体声或者环绕声道的平衡,所以要花不少时间寻找合适的地点、高度和角度。Schoeps CCM 或 CMC 系列话筒很难推,可能需要更高的增益。而 Neumann 145、DPA4000 系列、森海 8000 系列就很容易推,它们的增益就可以小很多。增益的控制不仅仅是为了音量,口子开得越大,进来的细节就会越多,而这种细微变化的小声音,是完全依赖细节的多少的。话放的品质、ADDA 的品质、甚至是话筒线的品质,都会直接影响最终录音的细节程度。

在底噪不到 35dB 声压的安静地方,增益必须是要开大的,这时候衣物、装备、呼吸声都会被话筒捕捉到。可以开大耳机监听音量来测试。但是录音键按下之前,一定要把耳机挪远!耳机里的声音会透出来传回话筒的。在录一些超级大声的声音时候,也是如此。

耐心,耐心。不要轻易按下录音键,耐心找合适的地点、角度个高度。不是每个声音都有价值的,后期会浪费你太多时间。一旦发现一个合适的地点,不要只录一条。即使 10 分钟一条,也最好再稍微动一下位置、高度和角度,再录一条。后期多一个选择都能救人的。再说,很可能下半辈子不会再来这个地方了。


上次在去乌苏前,我的录音计划里有很长一份目标声源清单。然而到了地方之后,你的选择瞬间消失了。太多的不确定因素。例如水,夏季的山里一半都会有充沛的水量,因此可以录到多种形态的水。然而当水量太猛、又是一个 V 字形深谷的话,水声的回荡大幅度提升了音量,从而导致不同地点的水声差异变得很小,环境底噪变得很大,好尴尬。原本计划录一些石头在山谷里滚下的声音,也完全没有了机会,那必须要有一个干的河谷才行。而且这样的水声通过峡谷共鸣可以传得很远,要避开的唯一办法就是翻阅山谷两侧的山体,到它们的背面去,那会需要很长时间,很可能 8 小时的行程里有 7 个小时在爬山。这是远远超乎计划的行程。或许在江南地区,3 天的行程可以录到丰富的素材,但是在乌苏这样的地方要 5、6 天才行,除非做好在山上过夜的准备,那没有当地向导是很危险的。山路、即使是密林里的兽道,行进也是非常非常慢的,一小时走不了多远。但是秋冬季节,山谷下的水量可能会小很多很多,而风会变得很强、很冷。


时间,时间,时间。丛林的环境,每个 take 会需要比通常更长的时间,很多经验丰富的老人会建议一个 take 能多长就尽量多长,比如一下录个半小时。尤其是那么安静的地方,自然的声音变化非常缓慢,你也不能预测一下次远处的斑鸠啥时候会再叫两声。你能做的就是耐心等待,享受这样的自然气味和那份安静。这也是野外录音让给我带来的最大收获,在那样地方录音可以放空自己,感受这个世界。


​ • 合适的时间,出现在合适的地点。当地的气候、地形和植被情况,尽量多学习一些,然后再决定时间和地点。 ​ • 并不是每个 Take 都会完美的,任何失误都有可能彻底毁掉一段录音 ​ • 耐心,不要被新鲜感冲昏头脑,冷静判断录音对象、发现有价值的内容 ​ • 准备好比你预期长一倍甚至 2、3 倍的时间来录音

How To: Recording A Birdsong Podcast in the Field

SongBirding 的創作者用的就是一個普普通通的 Zoom H1N 就能創造出播放量那麼高的播客,太不可思議了!他的錄製完全是實地的,有自然的腳步聲,自己的旁白也是在當時就說,而不是後期重配的。錄音最重要的是耐心,等到最好的那一個片段。在剪輯中,他儘量不刪去鳥鳴閒的空白,因為這本來就是很慢的事情。

For years I’ve been recording birdsong as an amateur recordist. Until recently my setup was an Edutige EIM-001 plus an iPhone. The EIM-001 is omnidirectional and high-gain, so it captures more sound, and in all directions.

Doing a podcast

Where normally the recordist aims to erase themselves from the recording, to keep the audio “clean” and without any hint of a human being behind a microphone, this podcast takes a more authentic approach to recorded birdsong. And it was so counter-intuitive that the idea took years to form.

That stated, once I got past the idea of not hiding myself, there were still many other challenges to deal with:

  1. Footsteps: gravel, grass, etc: It took a while for me to be comfortable with having walking sounds at all (first feedback I got was that it was actually nice and relaxing), but amongst those, I had to be careful around “noisier” terrain such as gravel, dead leaves, and very tall grasses. Interacting with these tend to make sounds that loudly cover many frequencies at once and can disrupt recordings a lot.
  2. Talking too loudly: No one but the mic needs to hear me, and if I’m too loud I’ll risk the recording hitting peak too much, so I tone my voice down a fair bit.
  3. Talking too quietly
  4. Wind & weather
  5. Distance & pitch (higher pitched birdsongs are quieter!): The one case where the mic doesn’t quite match my own ear’s abilities is higher pitched sounds. My guess is the windscreens might have some blame in this for absorbing some of the higher-pitched sounds a bit. I try now to make sure if there’s a high-pitched sound that I get really close to it.
  6. Anthropogenic noise: Airplanes, trains, cars, etc. Other people. Sometimes when the best birdsongs are being belted out, a jetliner flies low at the same time.
  7. Winter: clothing
  8. Insects: A later summer thing, though can happen any time when there’s mosquitos and blackflies. Those that land on the mic can usually be removed in editing, but when the cicada are out, they can dominate a recording with their wide-frequency sound.
  9. Patience: This one takes a while to learn, but listening to a bird sing for a whole minute can feel like far more time than that has passed, and it can be easy to underestimate how much time you’ve recorded something for. A good, clean recording can be a rare thing, but it is easy to forget that once you’ve had several iterations of the same birdsong over and over again. Always record more when you can, since you might not be able to get that same bird or species again that day or again that season!
  10. Flubbed takes: In a solo recording situation you are your own director, and while you can’t ask a bird to do another take you can do this to yourself by repeating something you feel you might have flubbed or not expressed correctly. You can always edit for the best take that way. On a couple occaisions I waited until a bird stopped vocalizing to re-take something I said earlier so that I’d have a clean recording. One could use this to correct uncertain bird IDs (multiple takes with various ID guesses) but usually I kept my uncertainty in to keep it authentic. If I ever did make use of multiple takes it would have been at times where being unclear might be a detriment to the listener.


Since recording in summer 2019 I’ve acquired a Zoom H1n, ~~which I now plug a Edutige ETM-001 into its Line In (the ETM-001 basically the same mic as the EIM-001 but with a different plug)~~⋯⋯

Bonus: Livestreaming!

My setup for this at the moment is as follows: the Edutige ETM-001 is plugged into the Zoom H1n. I then have a stereo cable running from the Line Out of the Zoom to the iPhone (using a headphone adapter). With this setup I can simultaneously record a session (the Zoom is recording) while outputting it to the iPhone to stream into Twitter. Since this is live and thus no chance to boost levels after the fact (except in a resulting recording) I tend to make sure the limiter is on (always a good idea anyways) and turn the gain all the way up.

Editing Into a Podcast

Birding by ear is generally a slow-paced thing, it would be a bit frustrating otherwise. For this reason I try not to edit out the space between an individual bird’s songs/calls unless it really begins to drag a lot, or if I’m editing a segment that is a feature on that particular song.

I try to set the pace of an episode in the first few seconds, much like one might set ground rules of a game at the start, so that the audience knows what to expect. Fast-paced birding-by-ear isn’t realistic and might get quite exhausting quickly. To match that, when using music I tend towards slower tempos.

I do some heavy editing between encounters of species – several minutes of quiet walking isn’t going to add anything to the podcast but time – but I do blend the editing in a way to sound continuous so things feel consistent, unless I specifically refer to a recording break when talking, in which case I’ll do a quick fade out/in.

Originally I aimed for 45-50 minutes per episode, but quickly found that around 15-30 is probably a better range. With the pacing being slow, anything significantly over 30 minutes might be too much. If I have 40 minutes of good material, then I have two 20-minute episodes.

Lavalier Microphone Techniques For Field Recording

Of all my microphones I probably use my lavaliers the most often. Whilst there are lots of good ones out there, like many recordists I invested in a pair of DPA 4060s. I find these tiny omnis to be pretty rugged microphones with crisp, balanced highs and extremely smooth, rounded lows. Since getting them I have been experimenting with various setups and recording techniques for stereo field recording.

Back on my first ever recording trip to Iceland in 2013, it was environmental recordist Chris Watson who first taught me that you can simply mount lavaliers to each end of a coat hanger. This provides a wonderfully straightforward, lightweight and cheap mount for a spaced pair. The hook means that you can hang it up or bury it in the ground. I still use the coat hanger when carrying my equipment for long periods such as multi-day hikes and when weight is the biggest factor. However it does have its limits. You can’t attach it to a boom pole or stand and some environments simply have nowhere to hang a coat hanger. It also makes you look like a ufo hunter from a low budget sci-fi. This can definitely cause unwanted attention in populated environments!

The Binaural Technique

Next I started using lavaliers for my audiobooks as a binaural setup. The recordings with the DPAs were way clearer than the other purpose made binaural mics on the market. We found that we got the most realistic results from literally taping the microphone to the auricle (by the earhole) This produced some really nice 3D binaural spatial effects. However I wouldn’t recommend it for use outside of the studio (or with anyone else’s microphones for that matter).

For stealth recording outside of the studio I’ve often used a quasi-binaural setup. I took a cheap pair of sports earphones, removed the speakers and melted a small hole in each earphone, just big enough for the capsule of the lavaliers. From there I attach a Bubblebee windshield and tape the cable to keep it all in place. A readymade solution would be something like BudFits. This technique allows you to record almost anywhere without anyone realising. I once managed to make a recording in the Panthéon, Paris, only for security to tell me I wasn’t allowed to listen to music! I still use this technique often as it requires the least amount of kit and is truly stealth. My biggest issue with this technique is that you capture every breath, every slight movement, even the sound of you gulping. To do a good job requires that you behave like a statue. Another drawback is that you can’t monitor your recording. I once did try bluetacking the lavaliers to the outside of a pair of normal earphones, but the results were not great. So in all some of the joy of recording is lost with this technique.

The Camera Arm Rig

Moving forward from the spaced pair coat hanger technique, I started using Colin Hunter’s setup for any environmental recordings. The lavaliers are on two flexible camera arms, attached to a Rycote portable recorder suspension, which attaches to a stand. The suspension removes all handling noise, and having a stand means you can leave the mics out in the open at whatever height you please. The flexible arms also means you can adjust the distance of the spaced pair. Whats more you can fold them in, making for easier transportation. My hangups with this technique are that the camera arms are stiff to adjust and a bit flimsy. Its also not that compact and little heavy. I’m quite unorganised with cables so find myself getting tangled as well. Not to mention that you’ve now progressed to a high budget and frankly alarming ufo hunter – Definitely something to be avoided in public places. Colin has just worked out a V2 which uses the stubby Gorillapod arms instead and looks to be a much improved setup.

The Baffled Blimp

The blimp mounted technique is widely used with lavaliers. It might be just the right balance for compact spaced pair recording, especially in public places. Every time I’ve seen it used, its been in medium sized blimps, as distance between the two mics is vital to having a wide stereo image. However I only have the Rycote WS1. This is their smallest blimp and comes in at a total length of just 280mm. Whats more, the microphones would be mounted at the end cap intersection, leaving just 170mm between the mics. Whilst it comes down to taste, a typical AB pair will be spaced between 400mm and 600mm apart. In order to improve channel separation I wanted to try using a baffle, or Jecklin disk between the two mics. The idea is to isolate the mics and block sound coming from the opposite side. The principles for a traditional Jecklin require an acoustically absorbent material be placed between the two mics, which should be 165mm apart, so ideal for me! However the disk should be around 3 times bigger than my blimp’s diameter. Whats more, Jecklin later revised his original specifications. He concluded an optimal distance to be a much wider 360mm apart. I was therefore uncertain as to how my setup would perform. For the disk I used Plastazote foam which at 28mm is a little thicker than the typical disk. Unsure as how best to mount the mics I asked the Facebook Field Recording Group. It turns out everyone has their own method, from a wire strip of cut down coathanger that spans the length of the blimp to a wood dowel. My method is to simply use dental floss, which can be hooked around the end cap mounting clips. Not thick enough to block the caps, but strong enough to hold the mics with a little tape. The result is a windproof, lightweight, clean and compact setup which can be handheld, boom or stand mounted. Plus it doesn’t make you look quite so crazy!

The Blimp Test

Yesterday I took the rig out into the Toulousain sunshine to get an idea of how it sounded. Recoding train passes seemed a good subject as the left/right separation of a flypast is always revealing. Whilst I would have liked a second pair of DPAs to test it against, my only other field microphone to hand was the Rode NT4. The stereo mic has 2 cardioid capsules in an XY configuration. At least this would give some spatial comparison. The mics were aligned, with the Rode being just 50cm closer to the sound source. You can listen to the results (MP3) below.

In listening back, the results between the two mics seem almost incomparable – I asked myself if I had high-passed the Rode by mistake, as the low frequency difference was enormous. Its clear that the DPAs do excel with this kind of bass-driven sound. The omnis also naturally sounded much, much wider than the Rode. In hindsight I should have set the Rode further back behind the DPAs. The real question was whether the stereo image would be as natural as the XY arrangement. At this point I must say I’m impressed, to me it really did sound natural. The disk seems to be adding a great deal of separation, despite it being on the small side. If anything I felt that the disk was doing too good a job. The sound attenuated very rapidly on the blocked side once the train passed. I’ll be using this technique whilst recording the French city of Toulouse over the coming months, so will update this post with further analysis in different recording environments.

Other Techniques for Lavaliers

There’s still several techniques I haven’t used much. One of the best things about lavaliers is just how tiny they are. It means that you can place them in all sorts of nooks and crannies that we usually couldn’t get to. They are often used in film sound on vehicles to close-mic the engine, or the inside of the cab. I read somewhere that they were buried in the desert on the Mad Max set and driven over! In an environmental setting you can hide them under rocks and leaves to get up close and personal with wildlife. There’s also surround sound which is a whole other topic on its own. I’ve already seen 4 channel blimp mounted 4060s for 4.0 playback, and DPA have their own baffled 5.1 solution which is regularly used for live sports and pro game audio.

In all we can see just how useful lavs are for field recording. Inevitably the 4060’s self noise is higher than in many larger diaphragm mics, but for its price and sound quality (especially in the low end) its a tough one to beat. The lavs are a vital component of my field recording toolkit.

AF vs. RF, Sennheiser

Question: Can anybody give me a quick explanation of why RF modulated microphones are less susceptible to humidity problems than are AF microphones?

Answer: Basically, AF capacitor microphones use the capsule as a capacitor to store charge. With one fixed plate and the other free to vibrate in sympathy with the sound, the capacitance varies, and the charge moves in or out of the capsule accordingly. This is measured by the head preamplifier and an audio signal results. All well and good, but the capsule is inherently in a high impedance circuit — it has to sit there with stored charge until the diaphragm moves and any changes in the charge are perceived as audio. In a humid atmosphere the stored charge finds it easier to escape on water molecules in the air rather than through the input of the preamplifier, hence noisy and reduced output, and misery all round.

The RF system (as used in Sennheiser MKH microphones) uses the capsule as a tuning capacitor for an RF oscillator – which inherently employs it in a low impedance circuit where a high frequency signal is being passed through the capacitor all the time. Changes in capacitance (caused by sound moving the diaphragm) alter the resonant frequency of the circuit (circa 8MHz) and so its frequency becomes proportional to the audio signal. A simple RF demodulator restores the output to a conventional audio signal. More complex and slightly more fragile (the basic frequency is determined by a crystal), this system is highly immune to the effects of humidity and is thus the preferred design to be used out of doors (or when moving from outside to inside on a cold day!).

Microphone Techniques for 2.0 and 5.1 Ambience Recording



I’ll be really brief, Itaro. The two 8040s … two cardioids. Since you mentioned FX recording (presuming film and/or TV) … and not radio broadcast. I own a Soundfield and a Schoeps fig 8 for M/S and other uses, and granted its a much easier combo to squeeze into one windshield, but really I have found the usefulness (as a sound editor) of MS recordings (or their stereo aspect) quite limiting. I like recording techniques, I’m a fan of Blumlein (the man and the method) and the whole history of the Ambisonic scene, but for me, for FX recording … very rarely MS, and when I do, it’s for a reason: I’d rather record mono and just know I’ve captured a really good ideal on-axis frontal image!

The several reasons I’m interested in MS include: music ensembles, interior or round-the-mic; docu coverage for musical subjects; stereo radio coverage of musical or ambient situations requiring no-nasty-surprises folddown; custom multichannel reproduction situations such as gallery audio or AV art. (I’ll thank John here for having previously pointed me to Ambisonics developments made by Harpex B).

But MS is usually less useful than many expect for film FX (aside from the mono on-axis bit … but that’s mono, not the ‘stereo’ component). What I would NEVER recommend, and my real answer to your question, is to choose one quite specific and limited stereo technique over another (quite specific and limited etc etc). If you know that you are going to be required to record MS then by all means the tools of your trade are going to be an MS rig. If you know you’re needing to record ORTF, then two cardioids and a crossbar (or a fixed mount if you know you’ll be recording ORFT an awful lot over other twin cardioid patterns: personally, I don’t like to restrict my stereo angle and the ‘on/off axis sweet/sour spots’ of my mics to 110 degrees or 170 mm when 120 degrees and 140 mm might be better).

I WOULD recommend to try to put together a small palette of mics for a range of applications. Out of all of my mics, my pair of MKH 8040s and two of my DPA 4060s get used the most, so within that kind of budget I’d heartily recommend both those models (I presume you know the DPAs are small lavalier mics, omni and good sounding: they’re incredibly useful for all sorts of FX situations). If you need to record ‘stereo’ with a clear defined central channel you’ll need to either go two channel MS with a figure 8 and your choice of mid mic, or plump with three mics, LCR, and a recorder to accommodate: my experience is that a multichannel LCR recording is far more commonly required in film FX than an MS. Then again, what is often required is simply either a mono recording of an object or a (recording-angle assessed and chosen) LR stereo recording of an event or ambience.

So, back to the beginning, briefly:

  1. Two cardioids = very useful general all round tools for mono or LR stereo

  2. A K&M stereo bar - costing nearly nothing and allowing virtually all cardioid LR positions, not just ORTF

  3. Ingenuity for fitting said bar into big fat rycote (or just use furry balls)

  4. A figure of 8 capsule once the desire for MS reaches the demand, and when its cost is justified above all additional patterns in the arsenal such as hypercardioid, shotgun or omni …

Depending on what you end up recording lots of, the figure 8 might indeed be more useful than any of these and a very useful third microphone to buy - I’m not knocking it, just the idea that (film fx recording in mind) M-S could be considered to be a SUBSTITUTE for two cardioids rather than a technique chosen for a particular situation or result.

I expect the excellent null characteristics of the veritable figure of eight might now prove useful to repel the barrage of verbal abuse from our many fans of Mid Side here, so please just wait as I turn myself ninety degrees …

相反的意見認為 MS 被低估:

M/S recordings are IMHO underrated. I remember a seminar with Bruce Swedien twenty years ago, where he called M/S - “Maybe Stereo”.

If you’re trying to record a sound source or instrument that’s not physically too wide, using a stereo technique with one microphone pointing on-axis towards the sound source and another one picking up the diffuse field, with practically no phase issues, can’t be a bad thing. :-)

And you can manipulate the stereo width or “air” of the sound effect by just adding a bit of Side-information to the matrix. IMHO very handy for sound effects editing.

With X/Y stereo, the microphones are picking up maybe a wider stereo field, but none of them are pointed at the source, they’re 45 degrees off-axis and not always capturing the sound source to the best of their ability.

While I like recording instrument sections in ORTF stereo, the mono compatibility can get compromised and it’s not possible to manipulate the stereo width as much as with M/S or X/Y.

組裝好的 ORTF 話筒的問題:

I understand the attraction of a stereo mic for ease of use, but most stereo mics (except the expensive ones) lock you into one angle / spacing which may sound fabulous in some circumstances and mediocre to awful in others. One size does not fit all situations, and what will give you a great recording in a good concert hall may not be a good choice for at home and vice versa.

For instance, my favorite piano recordings by far are made with spaced omnis. I’m talking about a concert grand on full stick in a great hall where you want to capture the acoustic space properly as well as render the full tonal range of the instrument. But try that in a typical small home practice studio such as what I have, and you will likely end up with an awful recording swamped by an overabundance of early reflections.

Conversely, X/Y cardioids (such as what you get on cheap stereo mics) sound terribly narrow and flat in pretty much any classical concert recording situation. But in the small home studio situation where you want to minimize the sound of the too-small room in your recording, you can set your X/Y array quite close and get a pretty decent recording.

Now, if you get a pair of separate mics, that gives you the ability to alter the angle and spacing to your heart’s content, dialing in the amount of acoustic you want (or do not) want to capture. No stereo mic will ever give you that level of flexibility. If you don’t know all of the arrangements or what adjusting angles and spacing actually does, hang around here and you’ll learn a lot about those things. Once you learn more about this, you’ll quickly wish you had bought separate mics.

The best thing for home recording of a piano in this situation is probably a spaced pair of good figure-8 mics which will have deep nulls to cancel the ceiling and sidewall reflections in a home studio. Then when you are in a nice concert hall, you can align them together for a mid-side Blumlein recording and be able to manipulate the stereo image and direct / diffuse ratio in post. That’s the closest thing to an all-purpose mic setup that would work for you, but it’s also going to be the most expensive. A pair of Sennheiser MKH 30 is on my to-buy list when I have a spare $2500…

Sorry this turned into such a long post. Here’s the TL:DR: The Superlux S502 will probably sound great in a concert situation as long as it’s properly placed and you’re in a good hall. But it will probably not sound great in your home, since it’s going to grab too many early reflections from very close walls and ceilings.


通常整个录音通道的音量调整 有一个基本原则就是尽量不要去动数字音量的调整 而是使用模拟设备去调整 目的就是防止数字音量调整过程中 BIT 记录深度的丢失

请问高品质的声卡 AD 模块到底强在哪?

@ 我为 M 狂:prismsound 声卡录音试验,48k 比 44.1 听感略饱满,低频信息更完善,88.2 有类似提升,比 48k 还好,继续提升 96k 就听不出区别了,工作经历告诉我相信耳朵比相信参数和理论要靠谱得多。


西西 ,2006-08-11

许多小型录音棚原本都是民用的楼房改装成的。由于是以家庭楼房格局搞录音棚,所以你要选择好哪一间房做录音室,哪一间做控制室。重要的是要尽量远离噪音源 (如电梯等).同时,最好要紧挨着的两个房间。这样方便在隔墙上开一个窗,便于录音师和配音员的交流。这个窗要做成玻璃的,最好做三层,中间那层玻璃做成有一定斜度(20°左右),这样可以增强隔音效果,每一层都用玻璃胶密封起来,中间不要留下任何东西,它们会传导声音。



如果装修的房子隔音本就很好。那么您可节省一些钱。装修方法如下:在四周墙面打好距墙面 15cm 左右的龙骨,在龙骨上订上石膏板,再在石膏板上订上岩棉吸音板,如果不是顶楼,房顶部分也可这样做;如是顶楼,就不必了。地面,一般用粗糙点的地毯铺上就行了,如果担心楼下,可以做成悬浮的,避免振动。

如果装修的房子隔音不是很好。那么装修方法如下:先在四周墙面离墙 20cm 左右的位置多沏一层砖墙(可考虑在此墙中塞入岩棉),再在离墙 15cm 左右上龙骨,龙骨内塞岩棉,再在龙骨上订上石膏板,再在石膏板上订上岩棉吸音板。房顶部分也打龙骨,内塞岩棉,再上石膏板、吸音板。地面如果也有噪音源传上来,则一样地打上龙骨,内塞岩棉,外订结实的板子,再用粗糙点的地毯铺上。记得门最好是做成双扇的,每扇都要用软包包过。关起来越紧越好,注意时间,因为夏天装修的门到了冬天会萎缩,造成关不严的现象。


@ 生活就是一头驴,2020-09-25:

其实做全屋的隔音并不难。房间 6 面浮筑,与原结构全部是软接触,地板使用弹簧支撑,在上面做龙骨与原地板就是软接触。在这上面铺设水泥,再铺设地暖。等于做了一个新的楼板。墙面承重在铺设的地板上,并与原墙面软接触连接。天花板也是这么做。在全部门窗在原结构和浮筑结构上分别安装并使用气密门锁结构。全部电路管线使用软管接入,地暖使用防爆软管接入,中央空调及新风系统把室内主机移到室外,使用静音箱接入。这么做完的话基本可以达到 24 小时在家随便折腾,左邻右舍压根都不知道你在家干嘛呢。

按一平米 1 万算吧,有高有低看用什么了。