The Official Google Blog announced yesterday that Google Translate was getting more text-to-speech translation options. English and Haitian Creole were the initial languages, and French, Italian, German, Hindi, and Spanish were added a couple weeks ago (I musta missed that!)
Google Translate has added the speech synthesizer eSpeak, which is adding text-to-speech for Afrikaans, Albanian, Catalan, Chinese (Mandarin), Croatian, Czech, Danish, Dutch, Finnish, Greek, Hungarian, Icelandic, Indonesian, Latvian, Macedonian, Norwegian, Polish, Portuguese, Romanian, Russian, Serbian, Slovak, Swahili, Swedish, Turkish, Vietnamese and Welsh. (Does this strike anyone else as kind of an odd selection of languages? Where’s Japanese, for example? Why have Icelandic and not, say, Arabic or Hebrew? No offense intended to anybody’s language.)
You can try this for yourself. Google Translate’s URL is http://translate.google.com/, while if we wanted to get an English to Hungarian translation for good food, the URL would be http://translate.google.com/#en|hu|good%20food. (By the way, that URL is gorgeous. I love that structure.) There’s a little speaker icon by the translation; click on it and a rather tinny machine voice will tell you a jó étel. You can contribute a better translation, but that’s for text only, not speech.
What I wanted to do after listening to this was go to Forvo and see how the pronunciation compared to a human’s pronunciation, but while Forvo had plenty of Hungarian words I couldn’t find either of those particular Hungarian words. Machine translation always makes me a little nervous, while machine pronunciation makes me slightly less nervous but still concerned. Google’s making all these languages available is a huge step forward, but I wish I had something with which I could compare these translations…
Dang it! I missed the 2010 International Day for Migratory Birds. I had plans to put up my birdhouse with tinsel and little lights too. HALLMARK YOU BETRAYED ME!
Oh well, I can wait until next year, and in the meantime use a recently-announced tool to identify the birds that hang around in the backyard. Dendroica is available at http://www.natureinstruct.org/dendroica. This site does not have the most extensive number of details on each bird, but it’s easy to search and incredibly easy to browse a large number of birds at a time.
When you first visit the site you can choose being a visitor from Canada, the US, or Mexico. once you’ve chosen you’ll get a list of species available (in the US there are 642) and a search box for narrowing them down. Click on a bird in the search box and you’ll see a picture and hear an example of the bird’s song. (Very occasionally there is not a picture available for a bird.) Underneath the picture is a description of the birds’ song and in almost all cases links to hear more versions of the bird’s song and see more pictures. If you can’t think of what bird you want to hear/see there’s also a link to get a random bird from the list.
There’s no data about habitat, or range, or anything like that, but this is an incredibly easy site to browse. If you’re interested in sparrows, search for the word sparrow and you’ll get a list of 33 species through which you can easily browse, comparing pictures and songs. Most other bird sites I’ve used would require a lot of page reloads to go through a list of birds like this. Very nice.
Registration is not required, but if you DO register you’ll have the ability to contribute pictures and songs of your own, as well as take quizzes based on the birds you’re looking at, or create customized lists.
If you need a lot of scientific and habitat data about a particular bird, this site is not for you. But if you want to quickly get a bird song or photograph, or easily browse through lists of birds looking for whatever’s been ransacking your apple tree, this site is terrific! Recommended.
Thanks to Smashing Magazine for the pointer to Forvo, an online dictionary of pronunciation. The Web site at http://www.forvo.com/ says “We want to have all the words that exist in the world pronounced and recorded, including names.” A site like this is useful for someone like me, who learned all her vocabulary through reading and therefore can’t pronounce anything…
There are currently has over 557,000 words with over 479,000 pronunciations in 240 languages. The front page of the site shows you the language of the day; if it doesn’t happen to be the language you want, visit the languages page which will list all the languages available, from Afar to Zulu; some languages only have a few words of pronunciation available.
English, at the moment, has just over 56,000 words. The page for English lists the top words and words that need to be pronounced; you can also search for words. Words can have multiple pronunciations; Barack Obama, for example, has eight pronunciations on Forvo. (Sadly, one of them is not a proper pronunciation but an opinion.)
Each word has its own page; for example, epitome. The page includes the available pronunciations — in this case six — along with a map of where the speakers live (they’re from all over the world.) The pronunciation plays in the browser, after which, if you’re registered, you can vote it as a good or bad pronunciation. (Registration is free and has other benefits, such as the ability to download MP3 pronunciation files.) In this case two of the pronunciations were voted down; one sounded like a recording error and another was a pronunciation many would consider incorrect.
If you don’t find a pronunciation you consider appropriate or you just want to add your own, you can do that. (This also requires registration.) All recording is done in-browser; you cannot upload recordings to the site. Recordings are limited to 2.5 seconds.
The site does not censor what kind of words are available, so there are some things here you probably don’t want your kids listening to. And while I was going through the words I did find some shenanigans. But on the other hand there was a tremendous amount of useful content, and the community voting seems to make sure that the good stuff rises to the top. Recommended for people who need some help with their pronunciation and language nuts.
I saved this resource without noting who sent it to me — was it reader TS? JS? Anyway, it’s a fun one. Second Hand Songs is a database of songs originally recorded by one artist but then covered by another (and possibly several others.) According to the front page of the site at http://www.secondhandsongs.com/, there are currently 30542 works, 117622 performances, 2332 samples and 36664 artists on the site.
There’s a basic search for songs and artists, but a more advanced search for artists, mediums, labels, and even visual performances. If you’d rather browse you can look at recently-added items or newly-released items. I did a basic search for Fabulous Thunderbirds.
The results page I got was divided into several sections, including original performances, covers, an album list, and what I presume was a list of other soundtracks/albums that the artist’s work had appeared.
You can also look at individual songs. One of the songs the FT covered was Wrap it Up. I clicked on the song title and got information about who wrote the song and the original recording artist (Sam & Dave) as well as a list of who’s covered it since (FT, Buddy Miles, and Eurythmics.)
For a site that’s all about songs I’m a little surprised that there aren’t song clips or performance clips, but this seems to be mostly reference. I did find it useful, though; I looked up artists I liked and then saw who had covered their songs, if any. It wasn’t surprising to see that Tom’s Diner by Suzanne Vega had been covered twice, but Luka?… FIVE times?…
In my job I’ve been spending a lot of time looking up this or that application, figuring out if it would be best to do something with a Web-based or a desktop app, trying to balance out functionality and convenience. And periodically I’ve come across a site called Aviary, at http://aviary.com/.
Aviary had lots of mentions in a variety of places because it offers several different image editors. The tools available include an image editor (also a vector editor) as well as a color editor and even a screen capture tool.
I didn’t look deeply into Aviary because for my image editing needs, I use GIMP. And I love GIMP. But my attention was taken back to Aviary today when I read on its blog that it had released an audio editor. So I looked a little closer, and — helloooooo, Aviary.
All of the Aviary tools have bird names, so the audio editor’s called Myna. The audio editor allows you to import music in a variety of formats, or record your own music. If you’re creating music for noncommercial purposes, you can also use materials from APM’s Quantum Tracks library in your music creations.
Here’s a screenshot of what it looks like. If you’re using clips from Quantum Tracks (these clips covered a lot of ground, though I missed world fusion and international beats) you can just drag them and drop them in to place. Importing audio is slightly more complicated but not much. Myna’s home page has a demo but this app is not very complicated and I found I could figure out a lot just by playing with it.
Want to see what other people are using Myna for? You can take a listen at this Myna gallery. Here you’ll find lots of tunes by your fellow Aviary creators.
Aviary is free but of course there’s a paid option available if you want more stuff. Features for premium accounts include more tutorial options, private collaborating, and more private files. The premium accounts run you an extremely-reasonable $24.99 a year.
I have never found a audio app I really like — it seems like they’re either way too complicated or don’t have enough features. I am surprised to see that one I DO like might be Web-based. And discovering that, I’m going to have to take a closer look at the rest of Aviary.
The LoC has announced a new online exhibit from the American Folklife Center: “American English-Dialect Recordings: The Center for Applied Linguistics Collection”. This site has 59 audio recordings — 118 hours — covering North American English dialects. The recordings were made between 1941 and 1984, but most of them were made between 1968 and 1982. The collection’s available at http://memory.loc.gov/ammem/collections/linguistics/.
You can browse the collection by title, name, subject, or place. (There are only two main categories of place — Canada and North America — but there are lots of subcategory covering city and state.) I went and took a look at the offerings from Kentucky. There were several recordings available, some of them described as oral histories, some of them were monologues and some of them were conversations. In some cases the listings were transcriptions.
All the listings I looked at had brief descriptions of the conversations, the time the recording was made, information about the speaker, and a subject listing. All the listings also had audio available in three formats — Real, WAV, and MP3. I found a couple of the downloads a bit slow (I would rather download the recordings and listen to them that way than play them through the speaker) but once downloaded they were fascinating to listen to.
In the midst of all this talk about the price changes for iTunes, music costs at Amazon and Wal-Mart, etc., I thought it would be nice to cover the Free Music Archive, which recently launched in beta. A project of radio station WFMU, the Free Music Archive currently has over 5000 tracks and is available at http://freemusicarchive.org/ .
A lot of the front page is bloggish but look on the right hand nav for a list of recently-added songs (with RSS feed), a list of the most interesting songs (again with RSS feed) and a list of genres for browsing, from country to international to jazz to pop. The subgeneres on some of these listings were amazing — Sludge Rock? Post-Punk? No Wave?
At any rate song listings include artist, track, album, and genre. There’s a button for quick play of the song and another for quick download. You can register on the site but you don’t have to.
As for the music itself — well, it’s all over the map. Some of it I listened to and left after a few seconds, some I hung around for a few minutes, and some of it I quite liked and downloaded. I enjoyed Max Tundra, So Cow, Hayvanlar Alemi (Ever listened to guitars that just made you feel happy? “Bahar Patlatan”.), 8 Bit Weapon, and Edith Frost. I’m sure I’m not even scratching the surface here — there’s a lot of music I didn’t even get to because I wanted to finish this writeup.
Bands have their own pages, with biographical information, tour dates (for some), discography information, external links, registered users who are fans, etc. Some bands have just one song available while others have albums and albums and albums. (A button on the band page lets you play all the songs on the page; handy.) Songs also have their own pages; the page for Al Duvall’s “Poppycock & Tommyrot” contains language, bitrate, and genre information as well as the Creative Commons license under which you can use/reuse the music.
I could try to give you some kind of summary here, but I think the fact that it took me over two hours to do this entry because I kept finding more artists to listen to pretty much says it all. Take a day off from iTunes and come over here and explore. Recommended.