I am not going to find them — it is true that most if not all frequency responses graphs for microphones are made for 20 kHz, even for the H2n:

BUT H2n can capture frequencies above it, depending on mode.

For now, the only AI response that I got was from pal that told me to go to doctor regarding my hearing. I am not sure what do you mean about reading the book — I am not an engineer or someone like this, and I never wanted to be, although as I said I have knowledge about vocals processing which is not hard to acquire. Please do not get me wrong, I just want to make sure that I get quality device in the terms of samplerate — that may sound crude, but please look at my response below:

This is a Zoom H2n spectrogram of 48 kHz voice recording:

Spoiler

Is that SM7B spectrogram with same parameters?

Spoiler

 

Spoiler: Review

The Shure SM7B is one of the renowned classics; it can be heard on countless commercial CDs and is still used in the world's best recording studios today. You can find out why this is the case, what it has to offer, how this microphone sounds, and what we think of it in our Shure SM7B review.
With the Shure SM7B, I have a dynamic announcer microphone on my table that delivers a truly impressive sound and is designed to satisfy even the most discerning of listeners. It's described as "endfire," making it an excellent fit for more than just broadcast studios, and its design also makes it suitable for video productions. But isn't that said that large-diaphragm condenser microphones are better, and if so, why?

This question isn't easily answered because various factors determine whether a microphone is a good choice. Large-diaphragm condenser microphones are very sensitive and low-noise, require less amplification and therefore phantom power, but tend to have a less neutral frequency response. They sound direct, warmer, and sometimes give the voice a silky quality. Their high sensitivity allows them to capture everything in the room and beyond.

This includes not only distracting room reflections, but also the dreaded leaf blower outside the window or the noisy neighbors in a thinly sounding apartment building, which is why good room acoustics or even a recording booth are essential for good sound. Otherwise, the result would have to be edited, which, depending on the tools and skills, can also be detrimental to natural sound quality. As with photography, the basic principle is to start with the subject and avoid potential problems right from the recording stage. However, if you can't optimize the room acoustics or generally only record voices in noisy environments, there's really no alternative to dynamic speaker microphones.

The Shure SM7B tested in this case is significantly less sensitive to room reflections and distracting influences, but it also requires direct speech input and requires higher amplification and low-noise preamps. In terms of tuning, it can highlight speech well and is comparable to typical reportage microphones, which can be used outdoors in virtually any situation, regardless of wind and weather. After all, every take should be acoustically intelligible; some things can only be recorded once. Furthermore, not every speaker is familiar with audio technology, which is why solid tools for certain applications have their place.

Before testing, of course, I do some research, because my experience with Shure, in particular, is admittedly somewhat limited; I've never even owned the classic SM58. Apart from the dynamic capsules, the Shure SM7B and SM58 are not similar, even though the SM7B is said to be a staple in many studios, especially in America.

The fact that this isn't the case in Europe is probably due to the pricing policy. The SM7B currently costs around €360, and I seem to remember it being significantly more expensive a few years ago. Therefore, the high-quality workmanship, which certainly meets professional standards, is not surprising. The Shure SM7B is primarily marketed as a speaker microphone, but that doesn't mean it can't be used for vocals. The second windscreen does suggest that it can be used for more exuberant vocals. However, in this review, I'd like to focus on its core capabilities, partly due to my lack of vocal skills.

The Shure SM7B at a Glance

The massive microphone, which is firmly screwed into the side of the mount, reaches me, nestled in Styrofoam. The total weight of the solid microphone is 766 g, the body is 19 cm long, and the diameter without the mount is approximately 5.5 cm at the rear. The side screws and the mounting angle adjust the height accordingly, giving the overall construction a truly high-quality finish. Included in the package are a standard windscreen and another foam windscreen, type A7WS, which slightly alters the sound character and improves the already good insensitivity.

This is also available separately as a replacement. It seems the microphone has been designed for outdoor use, but I wouldn't use this windscreen in everyday life. The XLR cable exits at the bottom in front of the 5/8-inch tripod socket. The microphone and mount form an inseparable unit, so the jack and microphone are connected by a fixed cable. The jack facing downwards simplifies permanent mounting on a tabletop or upside down on a boom and ensures neat cable routing. Also included in the scope of delivery is a 3/8-inch thread reducer and a cover plate for the rear adjustment options.

The dynamic moving-coil capsule of the Shure SM7B has a cardioid polar pattern and reproduces the nominal frequency response from 50 Hz to 20 kHz, which is more than sufficient for speech. The sensitivity is specified as -59 dB, and the impedance as 150 ohms. The capsule is internally decoupled and shielded against electromagnetic interference and features an optional presence boost and bass cut.

The switches for these two functions are recessed on the rear and require a pointed object to operate. It's not entirely clear where the effect occurs; unfortunately, I can't measure it, and there's nothing in the technical specifications regarding this. From my listening impression, this microphone seems to be less linear and is tuned more towards the speech range. The bass roll-off apparently works up to over 200 Hz and doesn't seem as steep, but it does intervene effectively in the sound. It's primarily intended to reduce rumble, which works excellently, making the voice sound leaner as a result. The presence boost works somewhere in the 5 to 10 kHz range and gives the voice a bit more punch, which can be particularly useful for voiceover applications. In any case, the suitability of the settings depends heavily on the speaker and the voice, so experimentation is essential.

Once you've found your preferred parameters, the included plate can be mounted vandal-proof with the pre-mounted screws. This is certainly an advantage for fully professional use, but in general, I would have preferred the switches to be positioned further out. This makes it a bit fiddly for spontaneous experimentation, but then again, there aren't that many settings to adjust. The fact that the switches themselves are recessed already protects them from accidental adjustment.

The Shure SM7B in practice

As is typical with moving-coil microphones of this type, the microphone is spoken into the end of the tube. In practice, the cable routing proves to be useful, although my tripod isn't quite suitable for this type of microphone. Overall, I think it's a clever solution in terms of handling, so the microphone is also well-suited for the ISOVOX Mobile Vocal Booth V2, for example, and the cable can be routed directly out of the booth, resulting in a very tidy appearance. The fact that the connection is permanently attached to the microphone isn't a problem thanks to the strain relief. Firstly, you won't be removing the SM7B from the swivel mount anyway, and secondly, the cable is stiff enough and securely mounted in the body. You'd have to be extremely clumsy to damage it.

I prefer the SM7B's design for speaking rather than having to sit behind a dense windscreen. For normal applications, the slimmer windscreen is more than sufficient; it slides snugly over the tube and appears to be permanently attached to the microphone. The solid version is probably primarily intended for vocals or possible outdoor use. It's nice that it's included, though; some manufacturers charge extra for this luxury.

As already mentioned in the introduction, this also shows that more amplification is needed. My RODE NT1 is correspondingly more powerful on the AI-1 audio interface, which is why the signal-to-noise ratio of the Shure SM7B is somewhat lower. Theoretically, this is audible; in practice, it's absolutely unproblematic given the intended use, and the voice can cut through very well. However, it's likely that better amplification could get even more out of the SM7B.

In terms of sound, the microphone impresses me with its detailed, tonally well-tuned sound and ease of use when speaking directly into it. Even at a speaking distance of 10 to 20 cm, the volume drops off due to its insensitivity; otherwise, room reflections would be more noticeable. The voice always sounds quite even, even when you move slightly out of the microphone's focus. The side damping is also impressive, allowing multiple microphones to be used in parallel in a discussion, something that wouldn't work with large-diaphragm condenser microphones. In the following sound sample, I compared it with my own microphones.

The price may seem a bit high at first. However, when you consider the room optimization required for a RODE NT1, this is somewhat put into perspective. Conversely, a better preamp is advantageous; perhaps a tube can get even more out of the SM7B in terms of sound. Anyone who wants to produce podcasts with minimal effort and good speech intelligibility or who wants to dub videos should definitely take a closer look at the Shure SM7B.

Conclusion

The Shure SM7B is certainly not cheap, but it is a solid speaker microphone that can be particularly well-suited for use in rooms with unfavorable acoustics. The two settings for bass cut and presence boost on the back are useful, but a bit fiddly. The cover for the two switches can certainly be useful in individual cases, as can the second windscreen. The microphone clearly aims to meet fully professional standards, and certainly does.

Source: www.amazona.de/test-shure-sm7b-studiomikrofon/

 

The comment I was responding to was about dynamic microphones generally. not going over 20K; and how the relevance of the DAW session sample rate and the sample rate at the interface is not even related to capturing human vocals resulting quality. The comment I was replying to was absolutely useless snark. Session sample rates only needing to be 44.1 khz and not delivering better results because human threshold of hearing is 20khz has been disproven so many different ways it's not even worth mentioning.

Try to remember, the Shure SM7B is a dynamic cardioid microphone. You are comparing apples and oranges in this instance.

Spoiler

Zoom H2n Microphone Type:

• ✅ Type: Condenser microphone
  
  
• ✅ Capsule configuration:
  • 5 internal microphone capsules
    
    
  • Capable of multiple polar patterns

Supported Polar Patterns:

• XY Stereo
  
  
• Mid-Side (MS) Stereo
  
  
• 2-Channel Surround (Front and Rear)
  
  
• 4-Channel Surround (360° capture)

So it’s multi-pattern, but not cardioid-only, and certainly not dynamic.

"The book" he is probably referring to, or at least should be; would be Bobby Owsinski Recording Engineer's Handbook, or even the Mastering Engineer's Handbook; both Fifth Edition now. $50 bucks each. I have both on my bookshelf. I'm not exactly sure what he intends for you to find in either one of them that the AI has not read already. You should not put too much stock into people who tell you to go read a book who cannot figure out free file hosting and cry and bitch about stuff people share with them for free.

It seems like some people have a hard time with people who actually know how to write and proofread their own comments. Here is what a real AI response to your "48khz voice recording" looks like. Some of your questions about the H2N are actually related to why the H2 Essentials newer model is one that some field recording users like it for. It can record at 32bit float. Using one as a studio mic for vocals is not one of them. If you add a 1 to your model, you only get the flu.

Spoiler

ChatGPT said:
If you're seeing spectrogram activity around 48 kHz in a voice recording captured with a Zoom H2n, that’s unusual and likely caused by non-audio artifacts or aliasing, because:

Key Facts About the Zoom H2n:

• Maximum sample rate: 96 kHz (which means the theoretical maximum frequency capture is 48 kHz, due to the Nyquist limit).
  
  
• Human voice: Typically occupies up to 8–12 kHz, with sibilants peaking at around 15–20 kHz in rare cases.

So Why Would You See Signal Around 48 kHz?
✅ 1. Aliasing or Digital Artifacts

• If there's poor anti-aliasing filtering, high-frequency content (like harmonic distortion, RF interference, or even internal clock noise) could reflect near the Nyquist frequency (e.g., 48 kHz on a 96 kHz recording).
  
  
• This appears as unnatural spectral spikes or static “lines” at the top end of the spectrogram.

✅ 2. Internal Noise / RF Interference

• The H2n might be picking up RF noise from nearby electronics, which can show up as narrow bands around 48 kHz.
  
  
• The built-in electronics (clock oscillators, switching power supply noise) can also contribute spurious high-frequency tones.

✅ 3. Ultrasonic Artifacts

• Some non-voice sounds (like jingling keys, metal impacts, dog whistles) can briefly spike near or above 20 kHz.
  
  
• If you’re recording in 96 kHz mode, the mic might be sensitive enough to pick up some ultrasonic harmonics—but not voice alone.

✅ 4. Spectrogram Misread or Software Glitch

• Depending on the spectrogram software you’re using (e.g., Audacity, Izotope RX, Sonic Visualiser), settings like window size or FFT resolution might cause visual anomalies, especially at the very top of the spectrum.

What You Can Do to Verify:

1. Re-record a voice at 96 kHz in a silent room, then check the spectrogram.
   
   
2. Compare recordings made at 48 kHz vs 96 kHz—does the spike disappear?
   
   
3. Use notch filtering in an EQ to isolate or suppress the 48 kHz region and hear if there's any audible difference.
   
   
4. Try a different recorder or mic to see if it persists.

TL;DR:
If a Zoom H2n voice recording shows signal activity near 48 kHz, it's almost certainly not actual vocal content, but rather digital noise, aliasing, or ultrasonic interference. The human voice doesn't naturally produce those frequencies.

If you have a specific spectrogram image or audio clip, feel free to share it—I can help you analyze exactly what's going on.

And of course, since the big, evil AI asked me to look at the spectrogram you posted I uploaded a copy of it for you. Here is it's reply about it:

Spoiler

ChatGPT said:
Thanks for sharing the spectrogram image. Here's a breakdown of what I'm seeing:

Visual Analysis of the Spectrogram

• Frequency Range: The upper portion of the image is black, indicating no significant signal energy above a certain point—this likely reflects a sample rate of 96 kHz, where the Nyquist limit is 48 kHz.
  
  
• Visible Signal: The energy mostly lies below 15–20 kHz, with strong horizontal bands in the low-to-mid frequency range—consistent with human voice.
  
  
• High-Frequency Texture: There is a fine grid-like vertical striping pattern extending up near the top (~48 kHz). This is not typical of natural vocal content.

What That Likely Means
The activity near 48 kHz is probably electronic or digital interference, not voice. This could be caused by:

1. Clocking Noise / Internal Digital Artifacts:
   • The H2n’s ADC or internal circuitry may leak high-frequency tones near the Nyquist edge.
     
     
   • These appear as fine vertical lines spaced periodically.
2. Spectral Leakage or Windowing Effects:
   • Depending on your FFT window size and overlap settings, energy can “smear” vertically.
     
     
   • This is especially true near hard cutoffs like the Nyquist limit.
3. Aliasing from Ultrasonic Content:
   • If there's any strong source of ultrasound (like metal impacts, fluorescent lights, ultrasonic humidifiers), it could reflect into the upper spectrum.
4. Mic Capsule Resonance or Electrical Interference:
   • The mics themselves or nearby devices could radiate EMI that the H2n captures, especially in an unshielded recording setup.

✅ Conclusion
What you’re seeing in the top band (near 48 kHz) of the spectrogram is not actual vocal content. It’s almost certainly digital or electrical noise, potentially exacerbated by the recording environment or the Zoom H2n’s internal electronics.

Tips If You Want to Investigate Further

• Record in total silence, then zoom in on the top frequencies: do the lines still appear?
  
  
• Try battery vs USB power (if applicable)—sometimes USB introduces noise.
  
  
• Use a high-Q notch filter at ~47–48 kHz and re-export the spectrogram to compare.

If you'd like, I can generate a cleaned spectrogram or run a notch-filter test if you send the original audio file.

 

Last edited: May 31, 2025 at 5:22 AM

He was talking about recording at 96khz, which means genius, you've got a ceiling of 48khz with that fancy old Nyquist Theorem. That's 'basic digital audio 101' I learned about 30 years ago when the studio I was at swapped out the 1" 8 track machine for a pair of fancy new ADATs and a BRC. Maybe try to keep up one of these days.