Our last post was from almost 6 months ago! I tend to underestimate the amount of work it takes to put together a polished post. What ends up happening is I’ll have a general topic I want to write about and then I’ll spend HOURS reading all these extremely niche papers on acoustical properties of food before I write a single word.

I know I’m making excuses but I have been wanting to write about CRUNCHINESS for months. So let’s just freaking do this.

I’ll be real with you – I like to snack.

I don’t have a sweet tooth but I really like to eat crunchy things like potato chips. Potatoes are such a versatile vegetable and I admire how many different ways you can cook it. I originally wanted to write about frying french fries but then I discovered a paper that is more generally about the crunchiness of foods. To be exact, this paper is a literature review about how the sounds of eating affect our perception and experiences with food. The author, Dr. Charles Spence, cited 127 papers in this review!! In my quick search I discovered that he is famous for his research on the ‘sonic crisp’. This guy is my hero.

The paper is called “Eating with our ears: assessing the importance of the sounds of consumption on our perception and enjoyment of multisensory flavour experiences”. It was published in a journal called Flavour which I am sad to report has ceased publication.

This review covers so much ground and it blew my mind. Some of the research dates back to the late 70s. I think I could write a whole series about crunchiness after discovering this. 

The review starts off by explaining how sound is a ‘forgotten flavor’ in the research world and the general public.

Dr. Spence analyzed research articles and books on flavor from top food scientists and noted that sound was referenced less than 5% across the board which we would expect to be closer to 20% if equal weighting was given to each sense. He then cites a study where 140 food research scientists consistently rated sound as the lowest sensory attribute in regards to flavor.

The other main study he cites was with 80 regular consumers who were asked to rank the importance of each sense for a few different food categories. Surprisingly(?) regardless of the food category sound was the lowest rated attribute as well.

Although I am an audio snob, I never thought to consider sound as a way to describe flavor. Apparently the International Standards Organization (ISO) doesn’t either. Their definition of flavor is the

Complex combination of the olfactory, gustatory and trigeminal sensations perceived during tasting. The flavour may be influenced by tactile, thermal, painful and/or kinaesthetic effects.

Part of the issue (or gap) here is that most people are typically unaware of what they hear while eating. One could argue that the results were limited or not realistic due to the design of the questionnaires and the setup of the experiment. But fortunately, there are a lot of smart researchers and more recent studies have revealed how sound affects our experience with food. Dr. Spence covers a lot of ground in this review as I mentioned before. I am just going to pick a few of the examples he described that I thought were fun. 

The first example is with Magnum Chocolate Bars. Their research team was looking to address some customer complaints on how to get the ice cream to stick better to the outer layer of chocolate. A new formulation was developed to address the concern but as a result the bar lost its distinctive cracking sound when biting into it and customers did not prefer the new version. Eventually the research team realized the cracking sound was a signature feature of the product and they reverted back to the original formulation. Magnum is projected to sell over $3 billion worth of ice cream bars this year and a quick peek at their website prominently features ‘cracking’ as a descriptor! 

Reading this immediately made me think of Kit-Kat chocolate bars and how the main points of their advertising is that you can ‘snap’ off a piece and ‘crunch’ into it. I found this ad on youtube and I can hear how the sounds are over emphasized. This is genius marketing. I’m getting hungry!

Thinking about candy, other favorites from my childhood include Poprocks and Crunch chocolate bars. For God’s sake, THE BAR IS CALLED ‘CRUNCH’. How have I never paid attention to this before? I feel like such a sheep. But also, take all my money!!!

Already this trek is starting to suggest how crunchiness is correlated to the pleasantness of a food but what about in regards to the quality of the food? The last major study I want to highlight from this review is work that Dr. Spence himself conducted back in 2004. This is the study on the ‘sonic crisp’. I’ve definitely been eating potato chips since 2004. I wish I could have participated in this. 

In the journey of the sonic crisp, Dr. Spence and his colleague had their participants eat 180 Pringles in a one hour session and rate the chips on their ‘freshness’ or ‘crispiness’. The participants wore headphones while eating the chips and during the session a few conditions were randomized on a chip-by-chip basis.

The chip eating conditions were:

• passthrough (no modification)
• overall attenuation by 20 or 40dB
• attenuation or boosted HF content (>2khz) by 12dB

One can of Pringles was used for each chip muncher but a third of participants felt that they thought the chips were being switched out during the experiment. 

The results showed that people thought the chips were significantly fresher and crispier when the overall sound level was increased and/or when the high-frequency chip crunching sounds were boosted. On the flip side, chips were rated softer and staler when attenuated. 

That sounds like such a fun experiment and the results make a lot of sense. The study is a great example of how sound affects our sense of taste. Just thinking back to my own experiences, I can definitely associate crunch with freshness like in regards to celery or apples. Haven’t you had an opened bag of hot cheetos lying around for a few days and you shamelessly take a bite of a stale, sad, lack-of-crunch cheeto only to be reminded that you should probably be an adult and eat something better for breakfast….?! No? 

Like I said earlier, part of the fun with snacking is having the texture of crunch and crisp. 

Watch this video of this precious French Bulldog eating some rice crackers. (sorry it’s not a pug. I am apologizing to myself because this is part of my aesthetic). Doesn’t this sound super satisfying?!

I hope this post made you think a little more about how we experience eating foods and how sound reveals so many interesting things about the world we live in. 

The literature review covers so many other interesting papers and topics. I can’t tell you enough how much of a golden (chicken) nugget it is so go check it out! What are your favorite crispy things to eat? Are there any tasty things you’d like us to explore? Let us know in the comments!

I think part of my writer’s block over the past couple of months is that I found a paper that covered so much interesting content and I wasn’t sure how to capture that in a digestible manner. Moving forward I think I may stick to more specific topics and get more into the meat and potatoes. I am still figuring out my flow and I appreciate you sticking along for the broil(?).

As a kid I remember going to the grocery store with my dad and seeing him knock on various watermelons to determine which one is the most ripe and optimal for eating. I never really understood the method to his madness. I would try to mimic his actions, but with my lack of patience I would just end up picking whichever one looked the most round and “tasty”.

Can you tell if a watermelon is tasty by knocking on it?

I have heard other people mention that there are tricks to determine the ripeness of a melon but I never thought to turn to academia for help. Wildly enough, there are actually quite a few papers about the acoustics of melons and the properties of their ripeness. Don’t believe me? Look at this, this, and this. 

Before diving into some research, I decided to conduct a mini interview with my dad to learn more about how he stumbled upon this. He told me that when he was a refugee in Pakistan he sold lots of watermelons. (He’s got a lot of stories as you can imagine.) His strategy was to bring the melon close to his ear and squeeze it to determine if you can hear it crack. According to him, the crack was a sign of the melon being ripe and a good indication that it would be tasty. He basically trained his ear to pick up on what’s tasty or not. Let’s see if science agrees with the street vendor. 

For today’s post the paper I am reviewing was published in the Personal and Ubiquitous Computing Journal and it came from a team at the Future Cities Laboratory in Singapore (fancyy). The title of the paper is pretty self explanatory: “Classifying watermelon ripeness by analyzing acoustic signals using mobile devices”. The research team developed an Android app to crowd-source data for identifying watermelon ripeness in real-time while also providing data for their classification model. The main motivation of the study was to help shoppers determine if a watermelon is ripe using the knocking method I have seen my dad use. 

One of the main factors that contribute to ripeness is the sugar content in the watermelons and the authors note that different sugar levels correlate to a different acoustic response. From the start the authors cite quite a few papers that analyze the acoustic impulse response of watermelons, which means my dad’s logic was already valid. Their general characterization of ripeness is that ripe watermelons sound dull while unripe melons have a tighter and more metallic sound. 

What does a tasty watermelon sound like? 

For those who don’t eat, breathe, and sleep acoustics, the terms “dull” and “metallic” may not mean much. To better describe what the authors are talking about I decided to consult YouTube for some examples. 

I watched like 10 watermelon videos and I found it interesting how people describe the sounds. It was definitely not consistent which I can see why this may seem like more of an art than a science. Here is a short video with a good recording of watermelons at different stages of ripeness.

In case you are a weirdo and have never eaten a watermelon before, the authors of the paper also included pictures of what the inside of a ripe vs unripe watermelon looks like. 

Ok, now that we have a baseline on what the heck sounds tasty, let’s dive more into the paper and the app they designed. 

How does the app work? 

The process of using the app is meant to mimic a real world situation in a grocery store. Therefore, users are asked to place their mobile device near the surface of the melon while knocking on it with their other hand. From there the app:

• Analyzes the acoustic response
• Discards the background noise
• Extracts the “knocking” event by detecting start and end points
• Classifies the result

The paper defines a few signal processing techniques to discard unwanted noise and extract the desired signal. A few of these parameters are commonly used in speech processing such as short-time energy (STE) and zero crossing rate (ZTR). For the “knocking” events, these signals can be broken down into 1800 samples (or snapshots) that are roughly 41ms long. (We are approaching math heavy territory but bear with me here.)

I am not going to cover all of their calculations but the STE measures the sum of the signal over time. If you think about this in regards to a speech segment, an “unvoiced” segment in speech will have a smaller STE. 

From there they define the sub-band STE ratio to be the ratio of the STE within intervals and the total energy in a frame. They defined their sub-bands into four time intervals: [0, T/8], [T/8, T/4], [T/4, T/2], [T/2, T]. 
Hang in there!! All of this explanation leads to this graph:

Here we can see that for unripe watermelons the STE mostly drops in the first sub-band and the STE ratios drop fast across the groups. However, for ripe watermelons the sub-band STE ratios are much more flat and consistent. I dragged you through all of this just so I could go back to what I said earlier: “The authors’ general characterization of ripeness is that ripe watermelons sound dull while unripe melons have a tighter and more metallic sound.” Now we have a nice visualization for that. Yay, math!

Maybe that was too much for what looks like a simple graph but now I am going to show you a really ~FANCY~ graph as a treat for reading this far. Another form of analysis the authors completed for their data set is plotting the spectrograms of the ripe and unripe melon signals. This is a pretty common type of analysis in the audio world. A spectrogram is a visual plot of the spectrum of frequencies in a signal over time. (Quick side note, if you want to have fun playing with spectrograms in real time, check out: https://musiclab.chromeexperiments.com/Spectrogram/) 

This graph is a great visual representation of how a ripe watermelon sounds “dull” versus how an unripe watermelon sounds “tight or metallic”. The bottom graph shows that the energy drop faster with the unripe watermelon which correlates to why it doesn’t sound as hollow or dull as a ripe watermelon. Visually we can see that the spectrograms are quite different which means that you can train a model to analyze recordings and, EVEN BETTER, you can train your ears to pick up on the differences. 

Moving on, the paper goes into more detail about the data set that was used for training and classification of the model. However, what they didn’t go into is what they did with all of those watermelons after testing… What I enjoyed the most is that through all of these signal processing techniques and calculations the authors were able to develop a model with an accuracy of 89.9%! Hell yeaaah. 

The authors also included a few screenshots of their app. It definitely gives me a bit of Silicon Valley’s “hotdog or not” vibes. Unfortunately I could not find the original app on the Google Play Store anymore. This paper was from 7 years ago so I can’t say I am too surprised. But I would be lying if I said I wasn’t a little sad.

Since I already started down this weird rabbit hole of melon acoustics, I decided to investigate what other melon ripeness apps are available to download. I had a bit of fun with this.

Melon apps? Huh…?

Discovery #1: Weird spelling of “Watermelon” with no downloads, a lot of math, and no knocking. Boring. *Next*

Discovery #2: Watermelon shooting! What does this have to do with ripeness? How dare you! *Next*

Discovery #3: Best & sweetest watermelon! Now we are talking. This is the closest app I could find to the one in the paper. It also asks the user to knock on watermelons and it analyzes the result. It seems fairly legit with over 100 downloads. The comic sans font is a little sketch but it seems to be the most legit from all the other apps I scrolled past. 

Bonus Discovery: This came up when I searched “watermelon ripeness” in the app store. It’s pretty irrelevant but I saw a pic of a pug so I took a screenshot. You’re welcome. Also, can we acknowledge that this app has over 10K downloads?! I clearly missed out on this ‘watermelon challenge’.

I hope you had fun geeking out with me today about the acoustics of watermelons. I enjoyed learning about how you can apply speech processing techniques and machine learning to characterizing watermelons. After reading way too much about watermelons, I feel like it is worth mentioning that there are other important factors that contribute to a tasty watermelon such as the density, color and appearance on the bottom of the melon, firmness, and sugar content.

However, I will take my dad’s tip that you can knock on melons to listen for what sounds tasty. I am looking forward to going to the grocery store and testing out my ear. I feel more confident that I can find the right melon that sounds tasty and I hope you do too. Cheers to eating tasty watermelons~

Honestly, it was great, and I ended up firming up my relationship with veggies. A trope in these types of diets seems to be replacing the space carbs would take up on your plate with vegetables. This means cooking and cutting a lot of veggies.

Little did I know… this also meant torturing a lot of veggies.

….huh?

According to a study from Tel-Aviv University, plants have been recorded emitting ultrasonic sounds when cut or left in drought conditions. This means that (take this logical leap with me, please) each vegetable I was mutilating all month may have been screaming out in pain the entire time. In other words, I learned I had (tomato) blood on my hands. I needed to get to the bottom of this.

Looking into the paper a bit, I read that researchers recorded sounds made by tomato and tobacco plants under normal conditions, after being left without water, and after being cut. The plants were left in an anechoic box and ultrasonic frequencies were recorded via two microphones placed about 10 centimeters away from the plants. Control plants were observed as well as plants that were cut or left unwatered. Below a simple depiction of the test setup (“acoustic” apparently means “anechoic” here…)

To determine how many times a tomato plants screamed under certain conditions, researchers counted the number of sounds emitted by the plants each hour, plotted below:

“Neighbor-Control” and “Self-Control” are sounds counted by other control plants and the same plant before it was cut or left unwatered (fact: tomatoes do have a “normal” sound they emit about once an hour). “Treated” represents the plants that were either cut or left unwatered. We can see many more noises emitted by these plants.

Show me the tomato scream already

Alright, alright, let’s cut to the chase. Ladies and gentlemen, behold: the sound of a tomato screaming

Though admittedly not the best graph, researchers say they measured these sounds at about 57 kHz at 60 dB SPL—wait a minute… we’re shown 60 dB… on the y-axis of a linear waveform?? The more I look the less sense it makes… so level is oscillating between +60 and -60 dB huh? These researchers have some explaining to do. To be honest this kind of puts me off of this paper a bit, but I’m too deep into writing this to turn around now!

Regardless of how loud this scream actually is, thankfully, 57 kHz is well above the highest frequency humans can hear. So we (I) don’t have to feel guilty about anything. Phew.

If a tomato screams and no one can hear it, does it actually scream?

While writing this I was concerned that maybe my dog could have heard all my poor vegetables screaming while being plucked or cut and I was very concerned. It was a massive relief to see that my dog couldn’t hear my tomatos being tortured either. But let’s just say that if you have a gerbil, you should keep them in another room while you cut veggies…

Looking into animal hearing frequency ranges, I realized that many other animals should in fact be able to hear these sounds. This begs the question: what would stop a benevolent animal from saving a tomato in distress? The authors discuss that animals may, in fact, have some incentive to intervene:

“…if plants emit sounds in response to a caterpillar attack, predators such as bats could use these sounds to detect attacked plants and prey on the herbivores, thus assisting the plant.“

As long as there is something in it for the bat I guess!

Okay, I’ll level with you… I WANT to hear the tomato scream

Well, fortunately or unfortunately for you, there are people out there that want to hear plants enough to have made an app for that. The world is a big place.

The inventors of PlantWave have apparently developed an app that allows plants to play music based on “slight electrical variations” over the surface of the plant. I’ve watched this video a few times on repeat and still am not sure if this is a late April fools joke or…

(My favorite quote: “With PlantWave, you’ll be able to wirelessly connect from your plant to your phone”. FINALLY.)

Now I’ll have to download this and cut some tomatoes up in real-time to see if I start hearing death metal… I’ll report back soon!

I have been so excited to formally start writing about this that I haven’t been able to sleep lately. Good thing I have my fancy drip coffee setup to help me get through this and life in quarantine…

Coffee culture is super fascinating to me. After learning more about roasting and brewing coffee, I realized that I know quite a few people who are very particular about HOW to make a cup of joe. I used to be one of those basic people that would only drink iced coffee because my teeth were too sensitive and I never experienced how different coffee can taste.

This is an interesting parallel to working in audio because often times I find that people don’t appreciate and understand the value of good sound until they have the opportunity to experience it!

After being exposed to the wonderful world of experimenting with brewing coffee, I have been fully converted. At work we have formed a mini coffee club where we keep a log of how we controlled our brew of the day and we come up with a plan for how we will change our recipe the next day. I love seeing a bunch of engineers over-engineer brewing coffee. We have a small manual burr grinder and my coworkers are verrrry particular about how we grind our beans. Since it takes a while to manually grind coffee I offered to purchase a cheap electric grinder. But people did not approve of my choice (another tangent I could write about but I will not go it into to spare you) so our middle ground has been 3D printing drill bits to attach a drill to our current grinder.

One of our colleagues has also been experimenting with roasting his own beans and we have been providing him notes about our preferences with his roasting style.

What makes coffee tasty?

There are quite a few variables that contribute to how to make tasty coffee such as:

• Where the beans are sourced
• How the beans are roasted
• How coarse or fine the beans are grounded
• Ratio of ground beans to water
• Temperature of water
• Type of filter used
• How the water is poured over the beans
• Total brew time
• and more…

For our first post, we are going to focus on how coffee beans are roasted.

More specifically, what are the cracking sounds made from the beans and how does this contribute to the overall taste?

We will be reviewing a famous paper published in the Journal of the Acoustical Society of America by Dr. Preston Wilson from the University of Texas at Austin. For a while, this study was the number one viewed paper for the Journal online, which was amazing due to the contrast of the simple natured content versus the serious and academic reputation of the journal. It was just really unexpected which made it fun and almost “viral” in the academic community.

This paper is publicly available and if you want to geek out more after this, then feel free to read the whole study !

Let’s dive in.

The paper begins by explaining that the range of roast levels (light to dark) are controlled by the roast time and temperature. The progression of the roast process can be monitored by what is defined as the “first crack” and “second crack”. These sounds are loud enough to be audible with no supplemental amplification.

The first crack occurs when the temperature of the beans are roughly at 200 degrees Celsius. Each bean has its own time to shine and experiences its own aural signature. The first crack settles after two minutes and the acoustic output drops until the beans reach 230 degrees Celsius triggering the second crack. During the second crack the combination of the gasses emitted plus the internal heat causes the beans to burn, which gives the beans an overall darker taste profile.

I love how the paper compares the first crack to the sound of popping popcorn while they describe the second crack as more of “Rice Krispies in milk” sensation. I don’t know about you, but that is already starting to sound tasty to me…(Should we write about Rice Krispies? Let us know.) It is important to note that the cracks have different acoustic properties and this paper provides a qualitative assessment of those differences!

The study was done with a consumer-grade drum-based coffee roaster with a common blend of Arabica and Robusta beans used for espresso. A digital audio recorder was placed near the roasting machine to record the sounds of the beans during the first and second cracks and the recordings were post-processed to quantify the characteristics of the cracking profiles.

The first part of their analysis included breaking down the distribution of the amplitudes of the cracks. In other words, they analyzed the peak of how loud each crack was over time.

From this graph we can see that there were a larger number of louder events during the first crack compared to the second. The author proposed that from this peak acoustic pressure data there can be a simple algorithm to acoustically differentiate the first crack from the second. Pretty cool! Already we have determined how to listen for the cracks in regards to loudness.

The second part of their analysis included analyzing the spectral differences between the cracks. 10 individual cracks were sampled at random and a fairly low resolution FFT was applied.

I love this graph because we have a very clear distinction of the spectral differences between the first and second cracks! The first crack has a peak around 800Hz and the second crack has a peak at around 15kHz. These peaks are in very different frequency bands which means that you can train your ear to pick up the audible differences. YOU can become the roasting whisperer of coffee! ?

The last bit of analysis the paper covers is looking at the differences between the rates of cracking. Timing is everything as you would expect and once again the characteristics of the cracks are different.

Here we can see that the first crack of the roast typically occurs within the first 400 to 600 seconds and the second crack happens around 620 seconds to 730 seconds. It is interesting to note that the first crack has more LOUDER occurrences although the second crack happens more frequently. Refer back to the first graph where there were 62 instances of the first crack and 241 instances of the second crack.

So now that we have geeked out a bit, let’s summarize what we have learned today.

This paper highlighted three acoustical properties of the first and second cracks for roasting coffee beans.

1. The first crack is audibly louder than the second crack by almost 15% in peak acoustic pressure.
2. The first crack is spectrally lower in frequency than the second crack (800Hz versus 15kHz).
3. The second crack proceeds at a higher rate than the first crack by a factor of 5.

COOL! So now you may be thinking, “What the hell? You just made me read all this random shit but how does this relate to taste??” Good question. Unfortunately this paper only covered a qualitative assessment of the physical roasting properties. I would have loved to see a second part of the study where they roasted beans with different controls (such as timing and number of cracks) and did a taste test to see what people prefer.

Coffee tasting is most definitely a THING. It is more formally known as “coffee cupping”. I have seen a few local coffee shops offer free cupping classes. It’s definitely a goal of mine to attend one of those. Maybe I’ll write a follow up post about it.

But, anyways, this leaves room for us all to experiment! Taste is obviously highly subjective and different people will prefer different types of roasts. Now that we understand how to listen for different acoustic properties we can better control the end result of our roast. This means we can learn how to LISTEN for tastier coffee beans. Once quarantine life is over and we can all get back to our “normal” schedules I think I will totally conduct some taste experiments in collaboration with my coworker that roasts his own beans. Maybe I’ll even get fancy and do some blind taste tests for my coworkers to investigate what are variables that our group thinks are tasty. Either way I’ll be sure to report back more of our results.

Sounds pretty tasty to me. Thanks for reading until the end and stay tuned for more!