Hello everyone,

I need some help (and I probably can help some here). I am having problems hearing some Japanese sounds. My problems are basically: some consonants sound like each other, like "ぐ" and "む", several times I don't hear the "ん" sound. Other times I can't separate the pronunciation of "橋" and "箸". And only sometimes I don't here a long vowel like "とう".

So, inspired by this post <http://fluent-forever.com/hungarian-stage-2-concrete-vocabulary/#.UwEDQ3mcVZM> I am creating a Anki deck to train my ears. But before I start bothering my Japanese friends to record the sounds for me, I need help to create a good list of pair of words to differentiate from each other.

So, can you guys help me to create this list? It would be good if we can use some frequent words.

Know writing this I can only remember:
箸　橋
直ぐ　住む
来て　切って

Do you remember having problems to differentiate sounds?

I know that the context can help to differentiate, but I want to learn to hear the sounds.

Thank you very much.

Sorry if this reply is OT:

I find using subs2srs and anki really good for telling apart Japanese sounds. Listen to the audio first, then look at the japanese subtitles and over time the subtleties are naturally learned.

Speakers very often pronounce things differently than the intended pronunciation anyway. Eg. In English 'batter' is often pronounced 'badder'
It seems maybe what you are doing will not account for natural variation between speakers' pronounciations...

I guess it would depend on your native language. I you have -g and -m in your language, with practice you'll also differenciate them in Japanese. For that audio with parallel texts is great (like in the core decks for example) and I'm not sure you need a deck just for the sounds in isolation.

For stuff like 来て/切って, and like always when studying sounds, if you know what you have to listen to you'll start to hear it. If you know one is "kite" and the other is "kit-te" (with a sort of stop/obstruction), you'll hear them eventually because they are really frequent (same with kita/kitta). Audio with parallel text should be enough here too (core again). Like until 6 months ago I had trouble hearing all the variants of the -ん sound, but then I looked them up (wikipedia) and started hearing them (like 関係 kang-kei that I was hearing kankei before).

As for 箸/橋 it's a much more difficult problem since it's related to pitch which is harder to grasp (especially if your native tongue is pitchless). Still, it's solved the same way: when you know one is hashi1 and the other is hashi2 (using a dictionary with pitch like sanseido), with practice you start to become aware of the difference. For more details I invite you to search all the threads about pitch on the forums.

Knowing actual vocabulary can help with distinguishing between sounds to a degree, since the syllables are contained within the word in general.

For me, I liken listening to Japanese to listening to things in my native language English. I don't necessarily distinguish between individual syllables when someone is speaking, but I've listened to enough English and I have a good enough vocabulary to make out what someone is saying without trying to pick apart each individual syllable in a word.

Thank you all very much. I will follow your suggestions.

I just noticed I can add information about the pronunciation (like pitch accent) on my deck (core2k/6k) using an Anki plugin. I will probably help a lot.

I honestly wouldn't worry too much about pitch accent. Discerning "橋" from "箸" would come from context (if you come across a side by side comparison they would be forcing their pronunciation of both words anyway. If anything, remember they usually say "お"箸). Besides its not like every person in Japan speaks with the same pitch so foreigners can understand them because their dictionary says the word is pitch1 or 2.

I grew up bilingual (second language learned starting at about 3-4 yo), and then studied three more languages by the time I was 15. So that might be the reason why, but I never understood how pronunciation is not easy and automatic.

I was in the same baffled state in class as well, whenever one of my (otherwise intelligent and hard working) classmates was butchering a foreign language.

But it's always been "how can someone not just repeat any sound or accented word they hear?". This is the first time I heard the notion of "not being able to hear it". Seems like if you can't hear it, you're pretty screwed. I wonder if it's in any way similar to not being able to fix tone deafness past the age of five?

If you're having trouble hearing sounds what you want to try and do is find minimal pairs, that is words that only differ by one sound. After you have a hand full of them the next step is to find more. Seriously, you going to want a crap ton of them. After you have a tone, go to forvo or rhinospike or something like that. Take the audio and make some anki notes. On one side of the note you should have just the sound and on the back the word and preferably the IPA for the word.

If you don't know the IPA learn it, its a pretty invaluable resource. I used the IPA to learn how to pronounce the sounds that don't exist in English like the ふ, where as the first time I heard it I heard an /f/ not an /ɸ/

Hi,

I'm insisting in this, so I'm creating a list of minimal pairs. Up to now I have this:

Pitch accent minimal pairs: https://docs.google.com/spreadsheet/ccc?...sp=sharing

Long vowels minimal pairs: https://docs.google.com/spreadsheet/ccc?...sp=sharing

Gemination minimal pairs: https://docs.google.com/spreadsheet/ccc?...naWc#gid=0

Moraic nasal minimal pairs: https://docs.google.com/spreadsheet/ccc?...tM3c#gid=0

Can someone, please, help me to improve this lists? I need to find more words with double consonants that sound alike.

Thank you very much.

In your Long vowels minimal pairs, おもう and うれい don't include any long vowels.
頂戴 and 長大 are homonimous in terms of both mora and pitch.

viharati Wrote:[..] おもう and うれい don't include any long vowels.

Just more evidence that I don't here the sounds. For me, おもう sounds exactly like おもお. =(

The amount of listening is also a factor and people argue about comprehensible vs incomprehensible input... Anyways I recommend not worrying about this so much since it is a matte of getting used to the sounds of the language which takes time. In real life you get context u dont have to be worried about hashi vs hashi

I'm still at a lower intermediate level, but just doing a lot of listening helps. The only thing I misheard for a long time was 了解！「りょうかい」; for the longest time, I kept hearing 「ようかい」。

From my experience to listening to Japanese, I also find it difficult to comprehend and distinguish sounds (it might be the speakers or headphones and grainy-ish audio), but when I learn new vocabulary, after a lot of listening exposure I find that while I still have difficulty comprehending particular sounds from words that are unfamiliar to me, I can almost immediately recognize words that I am familiar with.

And this is coming from having already listened to a consistent stream of 1500-2000 hours of Japanese through various sources. So even with that much listening experience and still not being able to easily comprehend sounds, I think it might be a difficult task in itself as a standalone studying method.

Now don't get me wrong, I can follow what I listen to in Japanese fairly easily and over a wide range of accents, but I still get sounds mixed up a lot.

I personally found that it's good to listen to a large consumption of Japanese audio for the benefit of learning vocab, because the exposure to that listening will make it a lot more easier for your sense of hearing to grab words out of a spoken language that seems to be a quick, long stream of sounds and syllables.