You Hum It, These Search Engines Claim They’ll Find It

Recently a couple of search engines have come to my attention that are designed to allow you to hum a few bars of a tune into them, and then they’ll tell you what the song is. These are Midomi and Humming Search from Nayio. That’s the promise; the reality was disappointing.

The concept is straightfoward; you allow the search engine(s) access to your microphone (either by dialogue box with Midomi or downloading a piece of software with Humming Search), and then you hum your tune. With Midomi you can hum for as long as you like, with Humming Search it’s a 15 second count down. The search engine then compares your humming to its database and provides you with a selection of possible matches. At least, that’s the theory of the thing.

Of course, the first problem is, ‘what exactly is a hum?’ Is it actually ‘hmmm hmmm hmmm’, or does it include ‘De da de deee dadada da’ as well? Is singing allowed? For someone who likes to know about search syntax, Boolean operators and so on, these are very important questions. Anyway, onto the searching. I hummed. I sang. I dah dah’d until I was blue in the face. I even got various songs from my collection, played them and put the microphone next to the speakers. In the end I even said ‘This song is…’ and named the piece(s) I was attempting to hum.

All to no avail. Humming Search would listen intently, go off and search and then – nothing. Not a suggestion, not an idea, not even an error message. Perhaps it just didn’t like my voice, or perhaps I wasn’t humming to the right pitch – the oral equivalent of getting my syntax wrong maybe? After warbling for 10 minutes or so I gave up in defeat. Clearly I am not destined for greatness in the field of musical entertainment.

Midomi was much more successful if you count getting results back as success. However, in that case I’d have to say that a search engine was successful if it gave me results about civil engineering when I’d run a search on cat breeding. True, I hummed and it listened. Then it gave me results, but they were not even close to my strangled renditions of various classics or popular hits. Eventually I realised that I had to admit defeat when my wife came into my study and said with a note of irritation ‘What IS that noise?’ To be fair, what response can one give? ‘I’m singing to my computer’ doesn’t sound too great. ‘I’m playing ‘what’s that tune’ with a search engine’ is little better. I did try ‘I’m seeing if a search engine knows the song that I’m humming to it’, to which she replied ‘If you already know what it is, why are you asking the question?’ And I suppose that is a basic flaw in this whole ‘you hum it’ system – if you don’t know what the song is in the first place, even when you are provided with a list of possibilities you’re going to have to wade through all of them to find the one you want.

So for once, I admit defeat. I was unable to get any kind of sense out of either search engine, and to be fair this could be entirely the fault of my singing/humming voice. Consequently I feel unable to recommend or scorn either search engine – you’ll simply have to try them yourself, but do it when no-one is around, because otherwise they’ll think you’ve finally flipped.

Opinions expressed in the article are those of the guest author and not necessarily Search Engine Land.

Related Topics: Channel: Consumer | Search Engines: Audio Search Engines

Sponsored


About The Author:

Connect with the author via:



SearchCap:

Get all the top search stories emailed daily!  

Share

Other ways to share:
 

Read before commenting! We welcome constructive comments and allow any that meet our common sense criteria. This means being respectful and polite to others. It means providing helpful information that contributes to a story or discussion. It means leaving links only that substantially add further to a discussion. Comments using foul language, being disrespectful to others or otherwise violating what we believe are common sense standards of discussion will be deleted. Comments may also be removed if they are posted from anonymous accounts. You can read more about our comments policy here.
  • madridguiri

    Midomi worked pretty well for me. I whistled. Without being boastful, I can safely say that my mediocre whistling is much better than the singing Danny does on his podcasts. It nailed all the current songs I did but faltered on the folk songs and golden oldies.

    Humming Search wouldn’t fire up for me.

  • http://cbmsnetworks.com Gene Linetsky

    Phil — your wife nailed it. Good post otherwise, too :)

    Query-by-Humming (QBH) is very difficult to do. As long as the number of indexed songs is very limited and everybody is searching for the most popular ones (just for the fun of it), it may appear to work. As soon as you try to index everything — and if you don’t, how can you promise to find that really obscure song that would merit this last, desperate attempt to remember its name after so many years — the recall and precision go to single digits and the whole thing falls apart. And this is even if you index the all the right things and don’t index the wrong things such as a nice guitar riff in the middle of the brige — unless this is exactly what someone remembers from the song, etc., etc. Oh yeah, and a great number of songs cannot be hummed at all.

    But here’s some good news: there doesn’t seem to be much demand for it. Midomi is already morphing into a MySpace clone. The only difference is that, in addition to the countless “this is me!” pics, the kids can now post the “this is me singing!” clips. It really has nothing to do with search. Boolean operators? Not quite yet.

    PS. Here are some of the earlier QBH attempts that are still online:
    http://www.musicline.de/de/melodiesuche/
    http://www.musipedia.org
    http://www.mu-seek.com/
    http://www.songtapper.com/
    http://themefinder.org/

Get Our News, Everywhere!

Daily Email:

Follow Search Engine Land on Twitter @sengineland Like Search Engine Land on Facebook Follow Search Engine Land on Google+ Get the Search Engine Land Feed Connect with Search Engine Land on LinkedIn Check out our Tumblr! See us on Pinterest

 
 

Click to watch SMX conference video

Join us at one of our SMX or MarTech events:

United States

Europe

Australia & China

Learn more about: SMX | MarTech


Free Daily Search News Recap!

SearchCap is a once-per-day newsletter update - sign up below and get the news delivered to you!

 


 

Search Engine Land Periodic Table of SEO Success Factors

Get Your Copy
Read The Full SEO Guide