Tokyo Polytechnic University Publishes Results of VOCALOID Study

What respondents who have listened to VOCALOID music before found appealing about it.  The reasons given are: genre diversity, ability to listen for free, singing that humans cannot achieve, VOCALOID characters, ability to find high quality songs and videos, ease of making derivative works, peer-to-peer creation, synthesis technology, sound quality, 3D hologram concerts, popularity amongst peers and other reasons, with the final column denoting people who did not find the music appealing.
What respondents who have listened to VOCALOID music before found appealing about it. The reasons given are: genre diversity, ability to listen for free, singing that humans cannot achieve, VOCALOID characters, ability to find high quality songs and videos, ease of making derivative works, peer-to-peer creation, synthesis technology, sound quality, 3D hologram concerts, popularity amongst peers and other reasons, with the final column denoting people who did not find the music appealing.
As part of Tokyo Polytechnic University’s ongoing activities related to Yamaha’s vocal synthesis technology VOCALOID, the university had recently conducted a study via mobile web surveys aimed at avid music listeners in Japan between the ages of 12 and 39. The survey asked the target demographic various VOCALOID-related questions, such as how they came to know about VOCALOID and what type of music they like to listen to. The university has now published the main results of the study, which involved a thousand valid responses, in a recent press release. Although the press release contains a plethora of graphs illustrating the data they’ve obtained, it also highlights some of the findings, including the fact that 40% of girls in their teens liked listening to VOCALOID music and that over 95% of all respondents recognized the name Hatsune Miku.

According to the press release, this is not the first time Tokyo Polytechnic University has been involved in VOCALOID-related activities. In April of 2012, the Department of Interactive Media (which focuses on peer-to-peer and participatory creation) from the Faculty of Arts acquired VOCALOID3 software for use in composition and song-writing practice. They have also invited PeperonP to give a one-hour public lecture on composing music with VOCALOID, which will be held this March 27. This particular study about VOCALOID aimed to probe views regarding and the current state of active music listening, and the results dealt with various aspects such as listening preferences, why listeners liked VOCALOID, etc.
Percentage of people in each demographic who like VOCALOID music.  The demographics are split across male (cyan) and female (pink) groups, with each group further split into age brackets of 12-19, 20-24, 25-29, 30-34 and 35-39.
Percentage of people in each demographic who like VOCALOID music. The demographics are split across male (cyan) and female (pink) groups, with each group further split into age brackets of 12-19, 20-24, 25-29, 30-34 and 35-39.

The first part of the survey dealt with how respondents listened to music. In one question, the survey asked people to select one or more genres of music that they enjoyed, with J-pop taking a significant lead at 83% of respondents, followed by anime songs (38.3%) and rock music (37.2%). Amongst the entire demographic, VOCALOID music factored in 8th at 17.4%. However, when broken down by age group and gender, 40% of females aged 12 through 19 liked listening to VOCALOID music, along with 28% of females in their early twenties and 20% of females in their late twenties. 21% of male respondents in their early twenties also reported liking VOCALOID music. When asked about music buying habits in the past year, 48.5% of respondents reported having bought music because their favorite artist(s) came out with a new release, 33.6% reported finding about music on TV or on the radio, 25.4% reported buying music after listening to samples on video sharing websites and 19.5% of respondents reported having bought music because they thought of some song they had listened to a lot before. Out of all the respondents, 19.8% reported not buying any music in the past year. The press release noted that among respondents between the ages of 12 and 19, video sharing websites and TV/radio were about even at 33.0% and 32.5% respectively. Finally, it concludes that video sharing websites have a big influence on avid music listeners since over 90% of respondents said they either enjoy video sharing websites somewhat or a lot.
Percentage of each age group who bought music in the last year due to various reasons.  Gray represents all ages and the other colors represent 12-19, 20-24, 25-29, 30-34 and 35-39 from left to right.  The reasons given are: new release from an artist they like, liking music from radio and TV, finding something they like on video sharing websites, buying songs they've previously listened to a lot and learning that some artist is famous (e.g. from music charts), with the last set for people who haven't purchased music in the past year.
Percentage of each age group who bought music in the last year due to various reasons. Gray represents all ages and the other colors represent 12-19, 20-24, 25-29, 30-34 and 35-39 from left to right. The reasons given are: new release from an artist they like, liking music from radio and TV, finding something they like on video sharing websites, buying songs they’ve previously listened to a lot and learning that some artist is famous (e.g. from music charts), with the last set for people who haven’t purchased music in the past year.

The survey also dealt with some more participatory aspects of music, with 45.4% of respondents saying they enjoyed performing music to some degree and 20.4% saying they enjoyed creating music to some degree. 23.2% of respondents have never performed any music and 62.9% have never created music before. Out of those who enjoyed creating music, 29.3% replied liking VOCALOID music, 26.6% enjoyed social media and 21.4% enjoyed browsing video sharing websites.
Breakdown of how much respondents enjoyed doing various activities, categorized into greatly enjoy (dark blue), somewhat enjoy (cyan), somewhat do not enjoy (yellow) and do not enjoy at all (orange), with the percentage of people who have never tried said activity in gray.  The activities are video sharing websites, anime, reading social media, posting via social media, playing computer games, playing music and creating music.
Breakdown of how much respondents enjoyed doing various activities, categorized into greatly enjoy (dark blue), somewhat enjoy (cyan), somewhat do not enjoy (yellow) and do not enjoy at all (orange), with the percentage of people who have never tried said activity in gray. The activities are video sharing websites, anime, reading social media, posting via social media, playing computer games, playing music and creating music.

The second part of the survey probed respondents about VOCALOID music, and started by asking them whether they recognize several related terms. For recognition questions in the survey, people were asked whether they knew what something was or if they have heard of something. 41.1% of respondents knew what the term ボーカロイド曲 (VOCALOID song/music) meant and another 26.5% had heard of it before. Other related terms such as ボカロP (VOCALOID producer), 同人音楽 (doujin music) and DTM (desktop music, i.e. making songs on a [home] computer) had 20.2%, 16.0% and 11.4% respectively of respondents who knew what those meant and an additional 12.4%, 18.3% and 15.3% respectively who have heard of the terms before.
Recognition rate of various VOCALOID-related terms: VOCALOID songs/music, VOCALOID producer, doujin software and DTM.  Percentage of people who know what the terms mean are indicated by the cyan blocks, with recognition by name only in lavender.
Recognition rate of various VOCALOID-related terms: VOCALOID songs/music, VOCALOID producer, doujin software and DTM. Percentage of people who know what the terms mean are indicated by the cyan blocks, with recognition by name only in lavender.

Following up, they asked the 676 respondents who recognized VOCALOID music several other questions. 63.8% of those people have listened to VOCALOID music before, with an additional 9.3% saying they plan on listening to VOCALOID music in the future. 10.8% reported having bought VOCALOID music before, with another 20.9% planning on buying in the future. 23.5% reported having sung VOCALOID music at karaoke, with 14.6% willing to try for the first time in the future. Lastly, 24.4% reported having played rhythm games featuring VOCALOID music, with 18.2% wanting to try in the future. When broken down by age group, 69.0% of respondents between the ages of 12-19 and 71.9% of respondents between the ages of 20 and 24 reported having listened to VOCALOID music before. Meanwhile, 12.1% of respondents between 30 and 34 and 15.5% of respondents between 35 and 39 reported having never listened to VOCALOID music before but would like to try in the future.
What respondents found appealing about VOCALOID music, with percentages segregated between people who like creating music (cyan), people who do not like creating music (green) and people who have never created music (yellow).
What respondents found appealing about VOCALOID music, with percentages segregated between people who like creating music (cyan), people who do not like creating music (green) and people who have never created music (yellow).

The 431 people who replied having listened to VOCALOID music before were then asked what about VOCALOID music made it attractive. 43.2% of respondents pointed out the diversity of genres, 41.8% liked being able to listen to it for free (on video sharing websites), 40.1% highlighted VOCALOID’s ability to produce vocals that human singers couldn’t, 39.2% liked the VOCALOID characters, 36.9% reported being able find high quality songs and videos (on video sharing websites), and 36.2% liked the fact that it was easy to make derivative works. The fact that music is being created by peers was liked by 32.9% of respondents and 32.7% were fascinated by the technology itself. Finally, the sound quality (24.8%), 3D holographic live concerts (17.4%) and the fact that people around them liked VOCALOID (13.2%) rounded out the major reasons. 15.8% of respondents who have listened to VOCALOID music before didn’t find anything captivating about it. For people who also enjoyed creating music, the percentage who liked VOCALOID for the diversity of genres, peer creation and synthesis technology saw big increases, with smaller increases in other categories. For example, the percentage of people who liked the characters only rose slightly when dealing with people who liked creating music.
Percentage breakdown of people who know each VOCALOID character (cyan) or have heard of the name before (lavender).
Percentage breakdown of people who know each VOCALOID character (cyan) or have heard of the name before (lavender).

The third part of the study dealt with VOCALOID characters. 56.4% of respondents knew who Hatsune Miku was, and another 38.4% had heard of the name before, with only 5.0% having never heard of Hatsune Miku. For Megpoid/GUMI, 15.1% recognized who she was and an additional 8.6% had heard of her. For IA it was 10.4% and 7.1%, and for Aoki Lapis it was 5.1% and 7.2%. The total number of people who had heard of the name Hatsune Miku before did not really vary amongst different parts of the demographic, but the percentage of people who knew who Hatsune Miku was did vary, whereas for the other charaters, rates for both types of recognition moved in tandem. In general, recognition rates were generally higher in the younger half of the demographic and were also higher for people who liked creating music. For example, 70.6% of respondents who liked creating music knew who Hatsune Miku was, and an additional 25.0% recognized her name. For Megpoid this was 28.9% and 9.3%, with IA at 21.6%/11.3% and Aoki Lapis at 13.7%/12.3%. Based on these results, the press release posits that perhaps one of these characters will become a “post Hatsune Miku”.
A graph showing the percentage of respondents first learning about Hatsune Miku through various mediums.  Video sharing websites and television corresponding to the leftmost bars dominate the rest.
A graph showing the percentage of respondents first learning about Hatsune Miku through various mediums. Video sharing websites and television corresponding to the leftmost bars dominate the rest.

The 950 respondents who recognized at least the name Hatsune Miku were then asked about where they first heard of her. 36.5% stated video sharing websites, while 34.2% stated they first heard of her on television. Social networks introduced Hatsune Miku to 6.2% of respondents, 2ch or its summary blogs introduced her to 6.1% of respondents and cross-marketing with businesses accounted for 4.4% of introductions. Word of mouth (3.6%), magazines (2.5%), video games (2.5%), news websites (0.8%), music CD stores (0.8%), radio (0.4%), comics (0.4%) and newspapers (0.1%) rounded out the rest, with 0.6% listing other reasons and 0.6% not remembering.
Percentage of people in each age group who first learned about Hatsune Miku via video sharing websites (blue bars) versus television (yellow line).  The age groups are 12-19, 20-24, 25-29, 30-34 and 35-39.
Percentage of people in each age group who first learned about Hatsune Miku via video sharing websites (blue bars) versus television (yellow line). The age groups are 12-19, 20-24, 25-29, 30-34 and 35-39.

For respondents between the ages of 12 and 19 and especially for respondents between the ages of 20 and 24, video sharing websites dominated, with 53.2% of respondents aged 20 through 24 having heard of Hatsune Miku from video sharing websites. However, for the 25-29 age group, the two options were pretty even, and television comes out on top for the even older age groups. The press release notes that Yamaha had previously done a survey in 2010 asking passers-by about VOCALOID, and at that time, only around 65% recognized Hatsune Miku, and amongst people in their 30s, not even half recognized the name. At that time, not even 10% of people had heard of Hatsune Miku through television or radio. It is the opinion of the press release writers that the increase in recognition of Miku stemmed from TV/newspaper reporting of the growth of user-generated content due to the explosion in activity of young people on video sharing websites.
Demographic breakdown illustrating the percentage of people in each category strongly wanting (cyan) to make a small donation to providers of free content, wouldn't mind donating (lavender) and would not like to donate (yellow).  Going down the rows, the demographics were: everyone, males, females, 12-19 year olds, 20-24 year olds, 25-29 year olds, 30-34 year olds, 35-39 year olds, people who enjoy VOCALOID music and people who do not enjoy VOCALOID music.
Demographic breakdown illustrating the percentage of people in each category strongly wanting (cyan) to make a small donation to providers of free content, wouldn’t mind donating (lavender) and would not like to donate (yellow). Going down the rows, the demographics were: everyone, males, females, 12-19 year olds, 20-24 year olds, 25-29 year olds, 30-34 year olds, 35-39 year olds, people who enjoy VOCALOID music and people who do not enjoy VOCALOID music.

The last part of the survey dealt with music production. The press release mentions people saying that with video sharing websites and social media, anyone was able to publish their own songs, and thus carrying the side effect of spread of free content on the Internet, with it getting harder to sell. Thus, one thing the survey wanted to find out was how consumers of free content felt about the producers of said content. To all respondents, they asked how they felt about giving small donations to providers of free content. Overall, 4.5% strongly felt they should donate and 52.0% felt it should be fine to donate. However, for people who liked listening to VOCALOID music, the percentages grew to 9.2% and 68.4% respectively.
Breakdown by demographic of responses regarding current (cyan) or future (green) usage of DAW/DTM software.  The bottom two rows correspond to those who like VOCALOID music and those who do not.
Breakdown by demographic of responses regarding current (cyan) or future (green) usage of DAW/DTM software. The bottom two rows correspond to those who like VOCALOID music and those who do not.

The press release also noted that VOCALOID software was originally software aimed at DTM users, and thus they were curious to see if the popularity of VOCALOID music has caused an increase in percentage of people involved in music production. First, they asked to see whether people have had experience with activities related to music production, or if they plan to try in the future. 7.9% of respondents have used DAW(digital audio workstation, used for creating music on the computer)/DTM software before and a further 23.3% were interested in trying. Meanwhile, 2.2% had used VOCALOID software before, with 25.1% interested in trying. There were other types of creative activities queried, including making ringtones, editing music on a mobile platform and making music as part of a video game, with percentages of 20.5%, 8.3%, 2.5% having done so before and 20.8%, 32.3% and 32.5% willing to try.

A breakdown across demographics showed that there was a much higher percentage of people who had used (20.7%) or were willing to try (34.5%) DAW/DTM software amongst people who liked VOCALOID music. This demographic also had a higher percentage of VOCALOID users (6.9%) or people willing to try (50.6%). Similar although not as drastic increases were seen in the other creative activities as well.

The press release concluded with words from the department’s Professor Kuhara Yasuo, who said:

Vocal synthesis has been an area of research for quite some time and is certainly not a novelty, but the social impact of VOCALOID is such that it is able to alter the structure of the music business. Anyone can freely upload or download music through video sharing websites, and this is how VOCALOID music spreads. Unlike restrictive content tied up by rights [e.g. copyright], this environment is friendly to viewers and to creators who can freely participate in the creation of derivative works, and creations get reused. Derivative works with titles such as “I made XX sing YY” and “I sang YY” motivate not only creators, but also viewers, and thus induce a creative chain reaction. Everyone likes listening to music, but the bar is higher if one wants to also compose. VOCALOID software, along with a sound bank library, allows the creation of a song by inputting lyrics and a melody. Although not everyone understands music theory, everyone can create and nurture a melody within themselves, express this melody using this new musical language called VOCALOID, and create a song in the process. It wouldn’t be surprising for VOCALOID music to be the impetus driving someone to start composing. I look forward to the future, when the creator demographic expands and all sorts of unique content gets created and spread.


11 thoughts on “Tokyo Polytechnic University Publishes Results of VOCALOID Study”

  1. I’ve always seen the concerts as somthing that would put people off vocaloid rather than draw them in. Nice to see how popular vocaloid is now though.

    1. Yeah, I’ve noticed that the concerts are something that western articles or media seem to latch on to as something strange, when actually the concerts are only a small part of what vocaloid actually is.

    1. Probably never. Vocaloid isn’t considered that interesting outside the fandom or when looked at considered weird or odd by main stream media.

      1. All it takes is for one celebrity to endorse it, and it’ll stop being weird. Who knows, maybe someday an important rapper will do a Miku collab or something. I know there are several who like anime and whatnot, and you never know what can happen in the music business.

        1. Yeeeaaahhhh I don’t see that happening. Miku has already been in the main stream media light before in the west. It didn’t go over so well.

          1. Meh that’s cause Toyota screwed up everything :). I want to see how well Miku do in Europe, now that would b an interesting sight to see.

  2. Thanks a lot, that was a very good synthesis.

    I did a bit a back enveloppe calculation: the age range polled at the scale of Japan represent a population of roughly 44.1M people. time 17.4% of people who like to listen to Vocaloid music that’s 7,7M.

    Given that the demographic polled was heavy music listeners, there’s something at odd with CDJournal’s article one year ago who was stating that around 8M Japanese were listening to Vocaloid Music. Because obviously on average there should be less than 7.7M people.

Leave a Reply to D K Cancel reply

Your email address will not be published. Required fields are marked *