Thankfully, their documentation included the connection diagram from the sensor to the Arduino Uno. It works; however, I think the sensor's sensitivity is lacking. I am planning on using it and applying it to the neck surface to measure the vibrations our neck produces every time we pronounce a consonant.
I saw a post on this forum that uses a transistor to amplify it but I don't know how to read the schematics properly. Additionally, it included a capacitor that I don't possess.
Thank you in advance for helping me out on this, hoping for your kind response.
Yes, it would be—if I were planning to capture sound waves. However, in my case, I intend to capture vibrations produced when we pronounce voiced consonants. These are the consonants that require our vocal cords to vibrate, unlike other consonants—take the consonant "S" as an example. When you pronounce "S", it still produces sound, but it does not require vocal cord vibration. I only need to record the ones that produce vibrations and the microphone can't capture such vibrations meticulously, which is why a vibration sensor is necessary.
The sensor can be used as a contact microphone, but the code in the link only takes 10 snapshots per second, which is not very useful for sound recording.
Remove the delay, use the serial plotter, and see what you get.
Adding analogReference(INTERNAL); to setup() will give you a 5x gain boost.
Leo..
You're right to question the use of the word "meticulously"—that may have been too strong on my part. What I meant to emphasize is that a vibration sensor, like a piezo disk, is better suited for detecting whether a person is actually engaging their vocal cords, which is critical for pronouncing voiced consonants correctly.
For example, try pronouncing the letter “B” in a whisper. You’ll notice it begins to sound more like a “P.” Even though both share the same articulatory shape, what sets them apart is vocal cord vibration. That’s the feature I want to capture—not just the audible sound, but the underlying phonation.
A microphone captures airborne sound waves, which is useful for many applications, but it isn’t as effective in detecting vocal cord engagement—especially in noisy environments or when the speech signal is weak. In contrast, a vibration sensor works through physical contact and picks up the mechanical vibrations of the neck, making it less sensitive to ambient noise and more reliable in distinguishing voiced from voiceless sounds based on actual phonatory activity.
This is particularly important for my undergraduate thesis on speech development, where I aim to analyze how well individuals—especially those in the Deaf community—engage their vocal cords during voiced consonant production. That’s why a vibration sensor is not just a guess, but a deliberate, research-based choice for the kind of data I need to collect.
Fundamentally there is no difference between a microphone and a piezo disk. They both pick up sound / vibrations. I think you are fooling yourself if you think there is.
I bet it won't fool your supervisor. Please ask him/her about this assumption you have made. I did have 21 years of supervising final year undergraduate student, often three or four a year.
I appreciate your input and the experience you bring—21 years of supervising undergraduate theses is certainly valuable, and I respect that. You're absolutely right that both microphones and piezoelectric sensors detect vibrations. However, the type and mode of vibration they are designed to capture do differ in ways that matter for my research.
Microphones are designed to pick up airborne acoustic waves. In contrast, piezo disk sensors detect mechanical vibrations transmitted through solid contact—like those on the neck's surface caused by vocal fold activity. This makes them more suitable for detecting whether vocal cords are actually engaging during phonation, particularly when working in environments with background noise or when vocal output is faint or whispered.
My thesis is not centered on capturing high-fidelity audio but on detecting vocal cord engagement during voiced consonant production. In this context, a contact-based sensor helps isolate the source of vibration more effectively than a microphone, which can easily pick up ambient noise and other unrelated signals. This is especially important when working with participants from the Deaf community, whose vocalizations may be soft or atypical but still physiologically engaged.
And yes, I’ve discussed this with my adviser, who is guiding me through the appropriate literature and methodology. My study also builds on existing research using vibration sensors in phonation and speech therapy, especially for cases involving Deaf or hard-of-hearing individuals, where subtle vocal fold vibrations may not translate clearly into sound.
That said, I’m always open to refining my approach, and I appreciate the push to stay critical and grounded in evidence.
I think what you are calling a microphone is not what I would call a microphone suitable for this project. I am not talking about a hand held microphone that you get say on a karaoke machine. I am talking about a contact microphone. These were often used in world war 2 for a navigator and pilot to talk to each other. These were known as "throat microphone", or contact microphones.
See:- Throat microphone
Or search for them and see who is eager to sell you one.
The output of these can be amplified like any other audio microphones if they are not sensitive enough for your purposes.
Thank you for the clarification—I now understand that you're referring to throat or contact microphones, not conventional air-conduction microphones. I did look into those after your reply, and you're absolutely right—they are indeed well-suited for detecting vocal fold vibrations directly from the neck.
However, after searching through available options, I found that most throat/contact microphones on the market are a bit beyond what I can afford as a student. Many of them are priced higher than my current budget allows, especially when factoring in shipping and additional components needed for interfacing.
That’s actually what led me to explore using piezoelectric vibration sensors—specifically analog piezo discs. They’re inexpensive, accessible, and easy to integrate with microcontrollers. Interestingly, I also found that many contact microphones on the market actually use piezoelectric materials at their core to convert mechanical vibrations into electrical signals. So in essence, the sensor I’m using operates on the same principle, just in a smaller and more budget-friendly form.
It might not offer the same level of comfort or signal conditioning as a commercial contact mic, but it does give me a way to detect vocal fold vibrations for the purposes of my thesis, especially with proper signal filtering and amplification.
It may not have the same level of polish or built-in signal processing as a commercial throat mic, but for my use case—detecting vocal fold engagement during voiced consonant production—it provides a feasible, low-cost solution that fits within the academic and financial constraints of an undergraduate research project.
I genuinely appreciate your insights; they’ve helped me understand the technology landscape better.
You can ask your supervisor if they can buy one of these for general lab work after your project. I know I would have been sympathetic to your request.
While doing my PhD research I had to put up a convincing argument to my supervisor to buy a 13 and a half bit A/D converter. It cost ÂŁ75.00 at the time, and while it might not sound a lot now, it was the equivalent of three months of my grant in the mid 1970's.
Yes, I do admit I used ChatGPT to help me write my response in post #7. I usually draft my thoughts first, but since I’m not that confident with my grammar and sometimes my ideas don’t come out as clearly as I intend—especially because I often write in another language first—I use ChatGPT to help proofread and organize my paragraphs.
That said, everything in post #7 still reflects what I genuinely wanted to express. I just needed help putting it into clearer and more structured English. At the end of the day, it’s a tool, and I’m using it to communicate better—especially on technical topics where clarity matters.
And just to be transparent, this post was also written with the help of ChatGPT.
I’m really sorry if it came across like I was cutting people off by using a tool like ChatGPT—that was never my intention. I truly appreciate the time and advice you gave, especially since you were the only one who directly addressed the core of my question and helped me understand things better.
I only used ChatGPT to help me communicate my thoughts more clearly, not to dismiss or replace the value of real conversations and insights from experienced people like you. Thank you again for your help—it meant a lot to me.
