This is a plan to invite people who are active in various fields related to sound and talk about various themes about "sound" by our chief scientist.
The first time, we talked to Dr. Kazuhiko Kawahara of the Kyushu University Graduate School of Arts and Engineering, which FINAL is currently conducting joint research. I wrote the text as it was, so I think there is something difficult to read, but I think that you can feel the words of the speaker from the text.
I will share the story that I heard for about 4 hours into the main items. Part 1 is "career".
Kyushu University Graduate School of Arts and Engineering Dr. Kazuhiko Kawahara
・ A career
・ A career
Hamasaki: Today, I will ask Mr. Kawahara to various stories. Could you introduce Kawahara -sensei first? thank you.
Kawahara: Call of Kyushu University. thank you. Born in the 1964 Olympics, the Tokyo Olympics before the Tokyo Olympics in 2021. I was born in Fukuoka Prefecture, but I have grew up almost in Fukuoka City. After going to a high school in Fukuoka City, Fukuoka Prefecture, I graduated from the acoustic design department of Kyushu University of Art and Design in 1987. I went to graduate school and completed a master's course for information transmission. The research theme is acoustic signal processing. I was in a digital signal processing laboratory. At that time, there was a normal noise canceling microphone that was introduced to Japan, which was being introduced in Japan, but I was studying equalization of speakers using adaptive signal processing. After leaving the graduate school, I joined the acoustic equipment manufacturer, and was related to the development of digital signal processing and the development of digital mixers. In 1991 and 1991, he returned as an assistant to Kyushu University of Art and Design, and is currently working as an associate professor. In the Japanese Society of Society, I have been promoting the chairman of the Acoustic Education Committee. There is also an educational committee called Education Committee, an audio engineering sosity, an AES, where he is the Vice Chair of Asia and Vice Chair. This is what my career is. While preparing for this interview, I wondered why I wanted to make a sound or was interested in the sound.
Hamasaki: Dr. Kawahara, I stopped in the middle of the story, but let me ask a little question before asking me about the reason I became interested in sound. In a graduate study of a graduate school, he performed digital signal processing adaptation signal processing, and in 1989 he was entered a sound equipment manufacturer and developed digital signal processing and digital mixing systems, but in this era. Was it a time when the signal processing began to progress, and practical use began to progress?
Kawahara: In the era, it seems that compact disks have begun to spread as media. Changed from the record to the CD. At this time, at this time, for adjustments and commercial equipment, it was not possible to calculate 100 steps with DSP (digital signal processing processor) for one sampling cycle. When I programmed the DSP chip, I was optimized by which commands and which commands could be executed at the same time. I myself, depending on the DSP, I did micr programming, another programming, and another way of implementing it. You can now handle the sound digital, for example, you can use EQ or use dedicated hardware in real time, but what you use for, and there is no such reliability. Can I use something? It seems that there was a story like that. Looking back later, what do you do if you stop on the way using a digital signal processing? Such a thing was really said. It was a time when analog was safe because it would change slowly, even if there was a problem. When it was digital, what I actually heard in the old tale seemed to freeze just before the actual performance. I'm afraid to use such equipment for production, so I can't use analog equipment only in SR (Sound ReinforceMent, a task that hears a lot of audiences to a lot of audiences in music concerts, etc.). Because it is a game, I think it was such an era, as if digital devices could be put in such places as effectors.
Hamasaki: In 1989, I personally remember well. Sony's 24 -truck digital tape recorder can now be used, carrying it from Japan to Munich, and at the State Opera in Munich, Wagner's Nibelung finger ring, an opera that takes about 12 hours to do it all. You recorded everything digital without using any analog. About 200 tapes. I recorded it on the tape and brought it back to Japan, but in the era of recording digitally on the tape, it was in an era when it was reliable. However, if the recorded sound can be edited on a personal computer like now, it is not such an era. Finally, at the time, AMS in the UK has come out with a digital audio editing device, a dedicated Digital audio signal, which can be said to be the pioneer of the current Digital Audio Workstation, and if it is a 2ch stereo digital audio signal, a little equalization or a pitch. I could extend the time axis or shrink without changing. Using this, I was able to manage my 12 -hour post -production for four months without a four -month break. It's such an era. At that time.
Kawahara: I think it was. Probably, I think that it was probably a challenge or a big adventure, such as Hamazaki -san recording in full digital like that, but it would be okay if it was all digital. Isn't it an era when there was such a challenger, the end of the 80's? Before that, digital equipment was mixed in some places, and the AD-DA was attached to both ends, so it was extra unstable. Under the belief that the digital mixer system we were involved in at the company, if all the leaders at the time were all digital, they would have to move, and they would move. Of course, we did various ideas for reliability. Probably, in the industry, the full -digital mixing system for live use, the SR system, was the first time in the world.
Hamasaki: That's right. Still analog. In this era, reliability, then good or bad sounds were discussed. Digital has a digital smell, or it looks like an analog. What about digital and analog? When I was told to give something like that, I was a little too far ahead. Probably, the digital mixing system with such a live performance, and doing everything with digital systems was extremely worldwide. But if someone doesn't do it, you won't know. It was an era where it was only discussed because it was analog or digital, so in that sense, 1989 90 years and 91 years, in 1989 90 years 91 years, the digitalization of professionals has progressed, and as you mentioned, CD. It is about to be widespread, and it is about how to use signal processing in earnest, but as mentioned earlier, it is a time when the DSP capacity of the digital signal processor is completely harder than now.
Kawahara: That's right. How different is it? I think it was only about a hundred -hundred performance, as if it were in the current smartphone. hard. However, we test what we can do in that restriction. In the company era, I read the DSP manual from corner to corner, this command and this command could be used together, and in fact, there were some restrictions to move multiple commands at once. I read from corner to corner. But I thought it was well thoughtful, certainly.
Hamazaki: When Kawahara -sensei was in college, was digital signal processing a curriculum?
Kawahara: That's right. I'm a 16th year student at Kyushu University of Art and Design, but it's the first grade in the undergraduate class called digital signal processing. I took it in the first half of the fourth grade. Right now, there is a digital signal processing class in the second half of the second grade. That digital is commonplace. At the time, when my signal processing class was my second year, I was learning the Fourier transformation and such a thing. In the first semester of the fourth grade, it seems like you can take it if you are interested. In the first half of the four years, a digital signal processing teacher said, "I will teach you for the first time in the undergraduate." Until then, there was no lesson called digital signal processing unless you go to graduate school. So, I think that digital signal processing has begun to be accepted, or that there was a revision that it would not be possible to teach as a university.
Hamazaki: You did research on equalization of speakers in a master's thesis. Was there any purpose that was chosen to be speakers? As an application of signal processing.
Kawahara: That's right. At the undergraduate instructor and the undergraduate instructor, the instructor of the undergraduate work was written in the list of graduation recruitment recruitment, such as an adaptive speaker research. what? When I asked my senior, I was told that I would adapt. I'm not sure, but I was interested in speakers and signal processing, so I was interested in speakers, and I wonder if I would do something good in digital signal processing. When I applied for the end of the third grade, I did not understand that the equalizer would be automatically designed, but I could understand that it was a research where signal processing and speakers were together, and applied. So I did it. Adjusting signal processing, in fact, the training was like a very specific field of conditions, but I do it while thinking about both speakers and signal processing. Especially equalizer. The speaker is a physical restriction that there is a diaphragm, so there are such problems, such as the fact that the DC cannot be regenerated, the vibration is divided, and the sound center changes depending on the frequency. I guess I understood somehow while proceeding with such research. So my understanding at that time was that far. After all, at this time, the master's research could only simulate. When you enter the company, if you design it hardware and write it in the DSP program, it may work in real time. Then I wanted to work hard. I thought that hardware technology was coming that far. If you are only satisfied and are satisfied, you would be useless, there is such a world. I was very excited to make a hardware and write a DSP program by myself, but it moves in real time, even if there are restrictions. In a super mini computer at the university, it was possible to calculate events for one second in one second, even though it took one second of events for one second. I was impressed with the DSP at that time.
Hamazaki: Is it an era where you couldn't do it until you could fold in real time?
Kawahara: Filters and devices for folding -in only have started. At that time, Motorola. I don't make DSP now and I don't even make MPs. Motorola was probably like Apple seen from Windows at the time, and the Motorola seen from Intel was a very sophisticated impression. It's Intel or Z80. Looking at the Z80 and H8 microcomputers, Motorola is very sophisticated. Motorola's DSP, DSP equipped with Steve Jobs's Next Computer. I wonder if it was the DSP56000 series. In fact, there are sister parts of digital filters dedicated to adaptation processing, combining Motorola's DSP and Motorola's adaptive signal processing filters to make my own research and development activity for my ambitions of my graduate school. I was doing it as part of. That's because it could be sold, so I could do it properly. However, if you put DSP and special digital filters densely, it will be very hot. I can't close the lid. Heat design is such a thing. And I remembered. At that time, one sampling cycle was thinking in NS units. The electrons are only 30 cm for 1NS, and they can only proceed with only 30 cm. Then, even if a certain part is pierced, for example, even if the logical value is changed from high to low, if the reaction speed of the parts is held, if you try to interact with the parts from the end of the base to the edge, the reply will not be in time. 。 If you think that the speed of light is actually fast, that's not the case. I realized that it was like the limit of such electronic speed. Because it wasn't really transmitted, as a programming technology at that time, if this part gave this kind of order and the other part of the opponent would not reply, it would not be in time. My senior told me. What is it? What? So please tell me how to use it like a circuit tracer. Actually, I reply, but I couldn't get it at the time or timing I wanted, so at this timing I had to wait one step properly. I felt something like the speed of electricity or the limit of light speed. I had such various experiences. I had various experiences at the sound equipment manufacturer, but digital, not only such work. When I entered the company, I came from Geisha University, and a person from another department in charge of the mechanism or the mechanism was told, "You're not good at electricity. Well, do your best." I myself was good at electricity, but I thought it was hard for electricity. Probably, I don't know it electricity, but I think there was already a reputation for those who knew what was different. So, thanks to the success of my seniors, I wonder if I was able to work hard at the company. After that, the company told me that there were a lot of technicians like craftsmen, so I was told that I would be trusted by that person. Even now, it's really impressive or trusted. It's really a chat, but in the middle of development, I was told to do a custom -made unit in the early stage of development, and I made an external part of one unit of 1U rack mount. The circuit was generally made by a senior, and I changed it while referring to it. The circuit was managed. Because it is a custom -made, it is necessary to make a hole in the panel to attach a power button, LED, pilot lamp, and a cannon connector behind. To make a hole, there was a prototype shop, or a person in charge of the prototype specialty, and I heard that the person was scary or a harsh person, so I wrote the drawings and panels I wrote myself while being quite scared. If you bring it with you, say, "I want you to make a prototype", look at my drawing, look sideways, and say, "It's made with such a drawing." When I said, "I wish I could open three holes," I was told, "Number of holes." So, return to your work. When I returned to my desk and said, "I got angry," I didn't know at the time, but the mechanical parts were mechanical parts, and whether to measure from this side of the panel like this. I have to measure it from this side and have such standards. I measured from the right, measured from the left, or was too free, and the next day, I rewrote the drawings and brought them, "I'm sorry. I rewrote yesterday's drawing. If you say this, you can make it, "I'll leave it." When I went to the dining room on the lunch break or something, I followed from behind, and said, "You're doing something interesting. Call me again." It's a bite. I thought I was surprised. That's probably my selfish and friendly interpretation. I thought he was a guy who could rewrite the drawing in one day. The acoustic students of the college student are not good at drafting or that. The experience that a real expert would help you will help you when you get over it, was done by a company employee. Great. Later, it may be because it was a college or acoustic design department in the development, but it was zipper noise. At that time, when I changed the gain of the zipper noise, digital mixer fader, and changed the gain for each sampling cycle, it was said that it was like a zipper noise. I have to smooth it. At that time, how late it was to be tolerated, at the time, a digital circuit, specializing in digital circuits, and a super engineer tuned in various ways. I happened to work a little overtime in the development room. I was playing with my experimental board, or playing the boards on the development theme. "Hey, Zipper noise Annen" says, "If you want to delay faders and gains, how long you want to delay it." I don't remember what Kansai dialect was, but I didn't know how much it was. I immediately said, "I think it's okay to 20ms." That person was wondering what I answered right away, or said, "You're stupid," but said, "Is that so?" I thought about the hearing effect of the hearing for a moment, and thought that if it was about 20ms, I would not be able to separate it, so I thought I would not understand. The next week, in a similar situation, I worked overtime with two people and said, "The last story." I didn't remember at all. "The last story, about 20ms is just right." If you say, "What happened to it?" He said, "Do all of the art and craftsmen know that?" Say, "Do you know?" I was wondering, "I know that kind of thing." I wondered if the arts design department of the Faculty of Arts and Engineering at the art craftsmen and the art engineering department had a characteristic because they were properly educated about the sound. If you think about it now, that exchange may be a part of the strengths and weaknesses of the people in the arts and crafts and sound design courses. It is also an strength, where such signal processing and hearing are connected properly. I didn't actually have such an experience for two years, but among them, I was able to study a lot in a lot of studies.
Hamasaki: Two years mean, the life of a company employee at the acoustic equipment manufacturer is two years.
Kawahara: That's right.
Hamasaki: In two years, I can return to college as an assistant. Did this seem to come back from the university, did you have any trigger?
Kawahara: Well, I'm very interested in returning to college and university. Probably, even if other universities were heard, the sound equipment manufacturer was a good environment, so I wouldn't have gone. I wanted to return to the unique environment of Kyushu University of the Arts and University, and I wanted to be involved in educational research. So it seemed to fit the human resources of the university.
Hamasaki: Good timing.
Kawahara: That's right.
>> Read more: Vol.1-2 "I was interested in sound"
