Is it faster to advance listening comprehension working through dialogue line-by-line for comprehension, or actively listening at full speed?

Sounds like you’re asking two things here - speed, vs. breaking down a large passage into smaller chunks. My recommendation is to do everything full speed as much as possible, particularly if you are a self-learner and/or outside of Japan. With that said, you can take an approach that gradually builds your ability.

I’ve recommended it before, but to say it again I am really a fan of Shadowing: Let’s Speak Japanese (Book + CD). It’s legit, at a native reading that I feel like was typical of what I heard when I was over there.

It starts N5 level and works up to N3 and N2 (there is a separate edition with N2 and N1). So the first sections are just two people talking, repeating single word sentences. (いい? うん、いいよ). Then it builds up to Speaker A says a single full sentence and Speaker B replies with a single sentence. Then A, B, A, B. By the end of the second volume you’re listening to a full speech given by a groom at his wedding.

So it’s native speed the whole time, but gradually works up in vocabulary, grammar, and length of dialog. Keeps things comprehensible and at a “N+1” level of difficult rather than N+10. And it works! I go back and listen to things I struggled to catch at first, and now it’s easy.

My process is as follows, when I have audio and a transcription and translation.

  1. Listen to the audio straight away and see if I can understand it as is, or get the gist
  2. Re-play a few times to see if I can pick up any more
  3. When I feel like I’ve extracted all I can, check the transcription and translation to see what I was missing
  4. Re-listen, knowing the transcription, until I can hear everything in the passage and keep it in memory in real time

Seems to be working for me so far.


I’ve only done full-speed listening because it’s all Drama CDs of various kinds or interviews with Japanese actors or some such.

I’d be too frustrated going at a slower pace and I think it’s more important to just let go of the fact that you don’t hear all the words or understand everything said. There are probably chances to correct your interpretations as the story moves along. But, if there is some obviously central word that I don’t get, I stop the recording and look it up (guessing the spelling) and then continue as before.

I find the natural talk the most helpful for hearing comprehension, because real Japanese people have dialects and unique ways to express themselves. So listening to actual conversation gives a really good sense of the range of how people talk and how you, yourself, might wanna express yourself.

But, real conversation is also quite hard in the beginning. The speaking tempo being higher than for a drama production, naturally, and people not bothering with perfect grammar, restarting sentences midway-through, interrupting each other or talking over each other etc. - it’s messy but so is reality.

For me, that sort of listening is still fun to do as it really builds confidence in listening when faced with less complex stuff - I don’t get as frustrated or as confused anymore when three people or more starts talking at the same time.

It’s like any listening skill - a matter of experience and repetition and you’ll learn how to compartmentalize what you’re hearing as you expose yourself to it more.


This is pretty close to what I have currently formulated based on this thread. It might be something like:

  1. Extensive
  2. Intensive
  3. Extensive


  1. Extensive
  2. Intensive
  3. Extensive, Extensive, Extensive

My other theories:
The ratio of extensive to intensive should change as experience increases

This is all very malleable though, since I only have so much to work with.

1 Like

To clarify, I’m saying that as they’re very different tasks that require different skill sets, using the same method for intensive listening as you use for intensive reading is silly, not that you shouldn’t be doing both.
E.g. trying to acquire new words and grammar makes sense while reading, but there’s so much other stuff to work out first with listening that I think it’s a poor use of practice time.

In which case, try to make what “listening practice” means as explicit as possible.

Write down exactly what you want out of it. Your goal is probably different to mine, so naturally what you should focus on is different.

Then write down all the processes involved in actually turning vibrating air into mental concepts, and all the skills required to achieve that. Then work out what you would need to address each of them, what’s common to all of them, what’s better practiced when reading/in lessons etc.

Google exists my friend, it shouldnt take too long finding studies about this.
And you will likely find better answers there than here regarding what is “technically” the best approach.

This is so unhelpful. Lots of really experienced people browsing here have studied this formally and can point to the new, best research. Or you can go to Google and get some uncurated studies out of context with no idea if they’re still taken seriously in language pedagogy now, never mind the endless parade of ads and scams Google lives to serve.


Stop listening at half speed. Listen at full speed.

I can attest to it definitely taking long to find anything useful. And then you have to work out what the hell they’re talking about in the paper, which often means you have to search for other papers…it’s probably a lot quicker (but still a slow process) if you are an expert.

Thats why you learn how to search properly and evaluate information rather than feasting on whichever pops up first :roll_eyes:
Its in no way more unhelpful than getting dozens of responses with no idea who is knowledgable on the subject.
Could you elaborate on who these “really experienced people” are for future reference ?

  • Even though there is some debate about the terminology, I’ll define “intensive listening” as “For each line, (1) listen to it once at full speed; (2) if I don’t understand 90 - 95% of it, read the line and look up all unknown vocab and grammar; repeat listening until 90 - 95% comprehension”.
  • I’ll define “active listening” as consciously listening with undivided attention.
  • I’ll define “extensive listening” as listening at full speed without breaks to check for comprehension.

My current working theory:

  • “Intensive listening” of input with low comprehensibility is better at advancing comprehensibility, whereas “extensive listening” of input with high comprehensibility is better at advancing comprehension. For example, let’s say that most input is only 50% comprehensible to me; doing lots of “intensive listening” may help make most input 90% comprehensible. That doesn’t mean I will comprehend 90% of most input, only that I have the “tools” to do so, which will allow me to get a lot more value out of extensive listening. (This makes a lot of intuitive because I have the same experience in English when I read some dense text about distributed systems or whatever.)
  • “Extensive listening” to input with low comprehensibility can advance comprehensibility–even to the point of fluency–but is a relatively inefficient way to do so.
  • Likewise, “intensive listening” to input with high comprehensibility can advance comprehension but is a relatively inefficient way to do so.

In my case:

  1. Extensive listening
  2. Intensive listening
  3. Extensive listening

seems like a reasonable strategy until interesting input becomes 80% comprehensible to me. Although the “intensive listening” portion technically involves a lot of reading, I can just make that my reading for now, especially since the content itself (Terrace House, in this case) is extremely relevant to my goals of becoming conversationally fluent (and the cultural context from the commentary doesn’t hurt).

Anyway, that’s my current theory. Let me know if you have any thoughts.

1 Like

As long as you are actively consuming it, I wouldn’t put too much weight on half or full speed in the beginning. I don’t think you’d necessary cover more ground at full speed since you’ll likely have to replay audio going at full speed to catch and understand everything. That said, I’d rather listen to something full speed twice or more than 50% speed once.

1 Like

Any podcast recommendations?

I think several people in this thread read this post as “50% speed” as opposed to “comprehension is 50% at full speed” so I re-worded it as “comprehension is 50% when listening at full speed”. I didn’t mean to slow down the audio, only stopping at almost every line.

I’d personally spend the intensive listening time doing more extensive listening/reading practice, but ultimately it’s your decision…

Just don’t spend more time optimising than doing and don’t optimise your routine into something you can’t live with.

1 Like

I agree in the sense that I think that’s the split I eventually want. It’s just that, right now, I can’t find any interesting listening material that is 80% comprehensible, so extensive listening seems to be relatively inefficient. I can’t prove that. That’s just the interpretation that makes the most intuitive sense given the information in this thread. I guess one more point, is that, because I’m doing spoken conversation for 1 hour, five days a week, I get a decent amount of extensive listening from that as well.

Personally I think the 80% thing is a myth…the hard bit when listening is recognising the sounds and the structures they form without having to devote any mental energy to it. I think that’s more to do with lots of exposure than trying to work out what the meaning they convey is.
Not sure where that fits into your model, but my point is that the hard work is making the mental shortcuts that mean half the sentences just collapses into a single unit. Then you have loads time to work out what it means without pausing.

Once you get to the point where you can do that reliably I do think it’s worth changing your approach slightly though…

1 Like

Interesting. This seems to be an excellent argument for more conversation practice with a tutor (with feedback) because it’s very good, from my experience, at building listening mental short cuts.

1 Like

ひいきびいき is an amazing show. Just a man and woman talking about things they like


Got it. I think if you need to stop each line to try to make sense of what you just heard, that’s fine. That’s part of what I mean by “actively” listening to something. If you are listening to something, understanding only some parts, and not bothering to figure out the rest, then you’re not actively learning. That’s the trap you don’t want to fall into, like I did. So to answer the question directly, I believe it is faster to advance your listening comprehension going line-by-line than blasting through the entire content.


OK, first of all, for listening, here’s my theory. It’s not based on research, just on my experience:
I think listening at full speed while still trying to understand whatever you can trains your ear to pick apart the language on the fly. It teaches you how to break down what you hear. Studying words though, especially while playing the audio again, teaches you what to look out for, and helps you recognise what you hear. Case in point: after months of listening to anime every day, and often falling asleep halfway through an episode (that wasn’t planned… I was just trying to squeeze a bit of anime into my schedule), I started the Tobira textbook, which was made for the old JLPT 2 (current JLPT N3-N2). I found myself very often able to follow what was going on in a text by listening to recordings even when my eyes and mouth couldn’t recognise and pronounce the words on the page as quickly. Also, I found that I was able to break down the syllables I heard quite accurately even if I didn’t know what they meant. As such… I guess I’d say both ‘extensive listening’ and ‘intensive listening’ are important, because it seems they train different skills.

In my opinion, anything that widens your vocabulary will improve your listening provided you know how those words are pronounced. In that sense, intensive listening is faster because it produces immediate results: by learning words that you couldn’t break down or couldn’t understand, you immediately increase your ability to understand those words. On the other hand, extensive listening may not widen your vocabulary, especially if your comprehension is low, but it may produce global results since it’ll probably make you better at deciphering words in general. I guess you should emphasise one or the other based on your current weaknesses. If your vocabulary is too small to understand more than, say, 10% of what you’re listening to, and you foresee yourself listening to similar content in future (e.g. you’re watching a drama series or you like gaming videos), it may be more profitable to do more intensive listening because you’re likely to hear the same words again. On the other hand, if your vocabulary seems fairly wide relative to the scope of your content, then extensive listening will be more productive.

Slightly off-topic comments on language learning in general

I personally don't do much research on 'the best' or 'the fastest/most efficient approach' because I found a language self-study publisher whose approach changed my life while I was learning French. Ever since, I've just tried to replicate their methods for other languages, because it worked for me.

The publisher is called Assimil, and their courses are used by Prof. Alexander Arguelles, an American linguistics professor who is fluent in at least 30 languages. He has a lot of videos on how he approaches language learning, so you might want to take a look. (NB: their Japanese course for English speakers only goes up to A2 in the CEFR. For French speakers, it goes up to B2. No advanced courses right now.)

Anyway, I’m not here to advertise, but I just wanna give you an idea of what their course did for me, and then I’ll explain their approach. You can take what you want from it. Before I started their course, my level in French was B1-B2 (i.e. intermediate learner fluent enough for rudimentary daily conversation and reading simple texts, but unable to handle newspapers or technical vocabulary. Also completely unable to handle full-speed native speech.) I was doing French as a school subject with other students around the same level as me. When I finally got their course, it took me about 3 months to finish, but even after just one month, advanced French was easy. Everyone else was struggling because we were trying to transition from the B tier (intermediate, good enough for basic general daily conversation) to the C tier (advanced, able to read newspapers and handle some technical vocabulary), and the problem with that is that the C1 and C2 levels are extremely broad in any language. You only get there by knowing lots of words from all major domains that the language deals with, which means tons of vocabulary acquisition.

So what’s their approach? Well, I’m having trouble replicating it perfectly on my own because they worked with parallel translations: target language on the left, translation on the right. You can’t always find those in real life. Also, they tend to break down sentences using additional literal translations. The idea is to give the learner natural text that could have been produced by a native, along with recordings, and then let him learn from context with explanations so the complicated stuff doesn’t make him get stuck. Of course, the texts are arranged so they get progressively more difficult, which is something we can’t always do ourselves. This is the way a typical lesson went for me:

  1. Listen to the text once without looking at it, just to get a feel of the language
  2. Listen again while reading to understand
  3. Read the text again while checking the explanations
  4. Listen a final time while trying to understand, and possibly while trying to read along at the same speed.
My approximation of their methods in real life, since I can't replicate them exactly. This focuses on what I do for listening. I'll briefly summarise what I do for reading as well afterwards.
  1. Watch an anime episode with English subtitles, only referring to them when needed and while trying to catch the words used.
  2. Watch it again while looking up words that interest me and possibly referring to a transcription
  3. Watch it again just for fun without subtitles

I’m currently somewhere in between steps 2 and 3 right now, because I’m watching without subtitles, but with a transcription somewhere else in case I can’t understand. Obviously, I don’t rewatch the entire episode three times in a row, because that gets boring, but I’ll probably come back to the same anime a while later.

At the same time, I try to find other material to read in order to continue widening my vocabulary. It can be a fairly advanced textbook like Tobira, or NHK news articles with help from TangoRisto or for kanji readings. That helps my listening too: every time I come back to a particular anime after a few weeks/months while doing other reading/watching, I find I understand more.

Finally, my main tool for language learning, concrete skill-focused methods aside: I read the dictionary a lot.

What allowed my fluency to skyrocket for French, aside from finding a really good, intuitive course, was that I aimed to transition from an EN-FR dictionary to a monolingual French dictionary ASAP. I’m currently doing the same for Japanese. The simple reason is that it forces you think in Japanese instead of in English while learning new words, and it gives you extra reading practice. Plus, if there’s a word you don’t know in the definition, you’re forced to learn it too, so you acquire vocabulary even faster. Also, most good dictionaries list expressions and common structures in the entries for each word, so you get to learn idioms and grammar too. I still use my EN-JP dictionary a fair bit (it contains more proverbs anyway), but I try to give the Japanese dictionary priority because of all these benefits.

(Not really an answer to the original question because I'm still working on improving 😊)

Oh wow another Assimil fan! The first book (passive phase) was my introduction to Japanese, and it was perfect (except I always had to hide the romaji under a bookmark or something). I studied it daily with my partner for 3 months, in the same way you did (listen hard, listen while reading, read) – but never got around to doing the second book (active phase) because we started taking lessons. I should really go through it at some point, the dialogs and little cartoons are quite funny.

For the curious

A spread from the first book

A spread from the second book

The way they split learning into passive and active phases made total sense to me, having naturally learnt French in this way (sadly without knowing about this method, just winging it during my first year living here).

When watching anime/dramas/movies in Japanese I take shortcuts though – because I’m lazy I only watch things once at full speed, with Japanese subtitles and usually no stopping.
I read the subtitles when I can’t understand much (doesn’t always help if I don’t already know the word though), and I only look up words that keep coming up that I can’t understand in context.

1 Like