Falar Ler Entender

Well well, here I am after falling off the planet for a four week hiatus! But I’m just going to go ahead and give my update as if this lapse didn’t take place. So here we go:

Things have been going well with my mission to learn Portuguese. The hardest part has continued to be the micro-goal setting and monitoring my progress. This combined with my never-ending concern that I might not be studying enough (perpetual college student complex?) leads to concerns I have to keep in check. For example, I might declare to myself “ughgh I did nothing today!” before realizing that, all small tasks combined, I was actually studying and/or practicing Portuguese for 2-3 hours.

yeah... boo hoo...

In this post, I’m going to focus mostly on using the free program Audacity to improve my comprehension and speaking.

This tactic was originally inspired by Idahosa Ness, of the Mimic Method. Idahosa uses Audacity to slow down songs, especially rap, in order to see which syllables in a language end up dropped from the “text equivalent” when speaking quickly. You can read a great guest post he wrote about this on Irish polyglot Benny Lewis’ blog Fluent in 3 Months. Since my last post, I’ve gotten in touch with Idahosa about his completed program The Flow of Spanish (which was really great to see) and have had an opportunity to help him beta-test his Flow of Portuguese program. If you’re in my position of wanting to improve Portuguese (especially accent and tone), you could always contact him and seeing if he can add you to the trials. It’s been a lot of fun and it’s great to have someone skilled in communicating about languages carefully review your audio submissions for sotaque advice. Just tell him a happy gringa sent you.

So onto Audacity. I used Audacity in college for some creative projects, but never thought of using it to analyze a language until I saw the suggestion on the Mimic Method website. Since then, I’ve been having a great time “capturing” random bits of Portuguese and putting them under the microscope with this software.

Here’s how it works: Download the free software and drag an MP3/wav of your favorite song or soundclip in to the editing window (if you only have a youtube video, you can strip the soundfile with this site). Then, under the Effect menu at the top of the screen, choose Change Tempo (not Change Speed)– this algorithm will slow the speech down while preserving the native tone of the language. I usually go for somewhere from 15-30% slower than the original recording.

a screencap of the free audio editing/recording software Audacity

Finally, put on your headphones, turn up the sound (LOUD), and start mimicking! You don’t have to think about the meaning or how you might write it just yet, just focus on hearing and recreating the sounds. Apparently, often when singing, people generally show reduced accents (or so says Susanna Zaraysky of Language is Music).

When I do this exercise, after concentrating on listening (to the sounds, not the words) for a few minutes,  I restart the file trying to repeat exactly what the artist says. I’m usually a second or two behind and sometimes if phrases are too complex,  I trip up and just jump back in whenever I can. Once I get a little more familiar with the material, I take one of my earbuds out so I can hear my voice better and start a mix of “parroting” (1-2 seconds behind) and synchronized singing. I like to have one earbud out so I can hear my own voice better, especially when I’m trying to sing along. I find this really helps me emulate the sounds of the artist, similar to the way a singer in a choir might calibrate their voice to their neighbor’s.

For my first song, I picked the song Pilotando o Bonde da Excursão by Marcelo D2. It’s mostly about getting stoned, but that’s okay because it’s upbeat, funky, and, most important: fast. It’s been about a month that I’ve been listening to this song at least 3-4 times several days a week and I deliberately only just looked at the lyrics today. Some sources I’ve read suggest that text can trip you up since it doesn’t give any indication which syllables disappear when language is spoken fast (Idahosa Ness calls this “The Flow”). Accordingly, I’ve been trying to focus mostly on the sounds, getting ready to understand the meaning later.

Marcelo D2 ("deh-doish"), a carioca rap artist

There’s one interesting theory I read (forgive me, I lost the reference…) that says you can’t necessarily hear/perceive a sound in a language well until you can make it. For this reason, often, I make a lot of crazy sounds while practicing. D2 says something like “Fala–” and suddenly I pause the recording and try out 5-10 different a sounds, repeating the ones that sound the closest and making all sorts of faces and droning noises to alternate between the open/closed back/front vowel scales I mentioned in this entry. It’s the kind of thing one might want to do behind closed doors…

After using this method on songs, I was watching my novela Vidas em Jogo about a week ago when it occurred to me– could I capture the sound of their rapid fire conversations and slow that down too? The answer, as you may have guessed, is yes.

For my “novela studies,” I generally pick either “listening” or “speaking” when I watch an episode, and either focus intensely on understanding the content or mimic the dialogue, losing sense of the words but focusing on making Portuguese sounds rapidly. (Riffing off Ness, again, who says some interesting things about “babbling” in another language in that blog entry.) By slowing down the novela audio just enough to free up some brain processing space, I found I could do both (understanding, speaking) at once, more or less.

In order to do this, I watched a scene on a youtube clip, trying to understand what the characters were saying. I watched it again mimicking their sounds. After I was familiar with the content, I took the audio from that clip and slowed it down by 30%. I was amazed to find just how many “hidden” words appeared as my brain started figuring out the vocal slurs on a slower timescale. I listened to this slowed down clip several times trying to focus on where one word became another, and recognizing the meanings. Then I started mimicking.

drama, drama... son character gets caught in a lie (Vidas em Jogo)

It was great to have the sound so slow because it gave me time to really adjust my vowels to line up with the speaker’s (the Portuguese a gets me every time…). Soon, I could mimick about 50-70% of the content on the 30% slower speed. I ramped it up to 15% slower and did the same thing. Finally, I set it back to normal speed. By that point, I could only mimic a few sentences reliably– but hey, those are full speed native Portuguese sentences!

The most important part of this exercise has been figuring out which parts of words might be likely to disappear when talking fast. For example, I’ve found that when one word ends in a vowel and the next word begins with a vowel, the second vowel usually “eats” the first (Eg. when said quickly, “minha irmã” sounds like “minhirmã” to me). I don’t claim to know how this works in my brain, but I’ve definitely noticed that working on mimicking small clips of speech has helped me with word transitions in general. When I read now, I somehow know much better than I did before how words should start fitting together. As a result I can read aloud more quickly and with more confidence. I’m definitely improving. It’s a slow process, but–as my former Portuguese teacher here said–baby steps, né?

One closing note that I found helped me: tone! I love the sing-song nature of Portuguese. I love the website I listed in this entry. Especially, check out this section on tone complete with audio files of typical tone structures in Brazilian Portuguese. If you pretend each one is a sound or a song and mimic it (the words/meaning are not important here), you can build up an arsenal of tone structures for easier, more authentic conversation.

And as a final thought, If you listen to a Brazilian news report or radio advertisement as an English speaker, you should immediately notice (read: be overwhelmed by) the enthusiastic up and down predictable sing-song. I’ve found that since coming here I’ve internalized this voice, especially one example in particular, a man, always incredibly energetic (if you’re watching Globo you’ll hear him say “No PROX-imo SAB-ado…! sing-song sing-song SING-song!”). Often, when I practice reading, I try to speak with this hyped-out singing reporter or advertiser tone. For some reason it makes it easier, and it also puts me in a better mood due to the energy it requires and my perpetual amusement at whatever the hell reason I’m doing it for (I don’t honestly know why, but my productivity skyrockets when I do this).

this is what I picture that completely insane guy looking like... oh wait that's Tom Cruise

So, in closing, I’ve been making a lot of progress thanks to the help of Brazilian rap music, my favorite novela, and the free software Audacity. I hope, if you haven’t used these methods yet, you might consider trying them out.

That’s it for now! And thanks for reading.

And those weekly micro-goals… right… someday…

Tags: audacity, audio, becoming fluent, brazil, brazilian portuguese, brazilian rap, conversational flow, falar ler entender, foreign language, frame of mind, goal setting, language acquisition, language goals, language learning, language learning goals, listening skills, mimic method, mini-quests, novelas, optimism, português, português brasileiro, portuguese, portuguese learning, pronunciation, rap music, rio de janeiro, second language business, second language for business, second language for work, setting goals, speaking skills, weekly goals