Home
Creators
Search Recent Missing Random
Posts
Search DMs Popular Hash Lookup Tags Random
Importer
Import FAQ
Register Favorites Keys Review DMs Logout
Community
Telegram ThePornDude
Home Creators Posts Import Register Favorites Logout

Episode 9: Pitch-accent patterns for nouns/compounds (Patreon)

Published:
2016-12-26 17:38:52
Imported:
2026-06
Tags:
haven't archived this post yet. have a subscription? use the importer!

Downloads

  • 448689465.mp4

Missing 1 full-res photo, 1 video, as this post has not yet been imported.

Content

Bibliography
Japanese Phonetics Index Page

Hello, Patrons!

In this episode I talk about the pitch-accent patterns for five mora nouns, as well as compounds. This one took longer than normal to produce—hope it proves to be beneficial! If you have any questions whatsoever please do not hesitate to let me know!

Thank you all for the continued support!

Dōgen

Files

Previews only

Comments

Jenny Brown

Aha. "Coffee words" are likely an exception because the Italian they are borrowing from typically stresses the penultimate syllable, which is what these are doing. I wonder if this is also the case for other Italian-origin loan words. It also seems like the longer kanji-based examples listed at the end are all dropping on the penultimate kanji in the compound rather than on a particular mora. I wonder if there's anything to that.

Dogen
>>84695657

Hi, Jay. JP here, Dōgen's business partner. I'm helping manage the messages here on Patreon. It might help to go over the lessons in the pronunciation sub-series that cover these characters (35-39 and 54) to better understand and familiarize yourself with the sounds. Otherwise it could just take a bit more time to take in and be able to catch these.

Dogen
>>101173144

Hi Jenny! There most likely is something to what you're noticing, but because I've personally found the guideline, 'listen for the 中高 pattern in compounds' so useful, despite it's somewhat vague nature, I've always just stuck with it. Best of luck with your future studies!

tensaimon

10:31 glad you included じこけんお, I'd really hate myself if i was saying it wrong...

tensaimon

7:32: Rule: Compounds need a downstep in the second element. - Pattern 1 (downstep on first mora of second element) occurs when there is no downstep in the second element as a stand-alone word i.e. compounds need a downstep, but there isn't one, so pattern 1 creates one! - Pattern 2 occurs when there IS a downstep in the second element as a stand-alone word i.e. compounds need a downstep but this time luckily there's already a downstep in the second element so that downstep stays the same and is used in the compound. (i think...corrections welcomed!)

Dogen
>>11324202

Looks good to me. That said, I wouldn't recommend memorizing this, as I mentioned in the video. It's better to know, IMHO, that compounds are almost always 中高, and to train your ear to listen for that ^^

Zion Gassner

Hi Dogen, After watching this video, I wanted to ask you this since I know you've studied linguistics. I'm a heritage speaker of Japanese, and I've noticed that sometimes I make mistakes that I want to call "compound-based pitch accent backformation," where I'll pronounce a word in isolation as 頭高 because it seems like it could be this way based on its pronunciation when in the latter half of a compound even though it really should be 平板. For example, I used to pronounce 大学 as 頭高 because of the common 「○○大学」 compound as displayed in this video. I'm curious if this sort of change has ever naturally occurred (to your knowledge) in Japanese, i.e., if there has ever been a word of one pitch accent type that has acquired a different acceptable pronunciation because of its primary usage in a compound which shifts the accent?

Thorn

Hi Dogen. hope you see this comment. I have been trying to research about the pitch patterns of onomatopeic words like ざあざあ or がらがら and it looks like they are all 頭高. Is this a generality ? Also, sorry if the topic has already been brought up.

Dogen
>>13043716

Hi Thorn! This seems like it's true, but it's not actually the case. For example, まじまじ and いきいき are both 中高. I do plan on covering this in the future however!

Crysenley

Hi Dogen, I finally subscribed to your Patreon today and am gratefully enjoying your pitch accent series! :) As for this lesson, love how straightforward the compound word rule is for native words (白人男性, etc.)! Compounds always confused me so much! However, I can't help but feel the rule you gave around the 7:30 mark regarding long 中高 words can perhaps be simplified a bit. All the examples with 頭高 components were also loanwords. So if anything, I feel the loanword rule precedes the long compound word rule, as mentioned in this video by Japanesy. <a href="https://youtu.be/CnBYZJ33MsY" rel="nofollow noopener" target="_blank">https://youtu.be/CnBYZJ33MsY</a> For 3 mora or longer foreign words, the accent always falls on the third to last mora--unless it's an elongated mora, in which case it'll skip to the original mora (コーヒー). So, I feel compound foreign words are treated as a single word, with the accent still falling on the third to last mora. So my actual question! ^^;; Can it be said that the rule for native compounds is almost always like 白人男性 while for loanwords it's just like any other loanword? Thus, loanwords can kinda be ignored as compound words?

Dogen
>>13124245

Hi Crysenley! Thank you for the support, really appreciate it! ^^ Please allow me to give some advice below! First, I would encourage you to be just a bit careful with the phrase 'For 3 mora or longer foreign words, the accent always falls on the third to last mora'; I assume you know that this isn't true, and that you meant to say 'The majority of 3 mora or longer foreign loan words'. Some common examples of words that don't follow this rule are パソコン, コンセント, アクセント, アメリカ, イングリッシュ、etc. It's useful to remember that just because the majority of a certain type of word follows a particular pattern, this doesn't necessarily mean that the majority of everyday words that we use in conversation follow this pattern as well. That said, I do find that the '3 mora from the end' rule for loan words is generally useful! As for your actual question, unfortunately this isn't the case. For example, the foreign word compound for American Indians (アメリカインディアン) and Native American (ネーティブアメリカン) (arbitrary examples) both have their downsteps earlier than 3rd to last mora. And, there are many Japanese words that don't follow the rule I introduced for 白人男性, where the mora comes at the beginning of the latter compound. For example, 経済学　（けいざいがく）, or economics has it's accent on ざ, which is still part of the first word, rather than coming at the beginning of the second word. Likewise, 教育心理学 （きょういくしんりがく）, or educational psychology, has it's downstep on the り, rather than at the beginning of the word 心理学. So there are in fact many rules for both Japanese compounds, foreign compounds, and mixed compounds, but they are all very difficult to remember and utilize in conversation, which is why I don't recommend trying to master them all and to instead go for the much broader 'listen for 中高 in compounds'. Again, I feel that if you're in it for the long-haul this is the best advice, because if you keep this in mind you'll slowly but surely begin to develop a sixth sense for exactly where the accent goes in compounds, regardless of their composition. Does this answer your question? If anything wasn't clear please let me know! Cheers! ^^

Dogen
>>125461645

(Kimi and JP are helping Dōgen respond to comments and questions.) Hi, Zion! I tried to think of compound words where the accent of latter half is altered: 山田高校, 推薦入学, 株式会社, 洗濯洗剤, コーヒー牛乳, 風呂吹き大根, 救急病院, 車内広告, 花嫁修行, 卒業旅行, 東名高速, 別府温泉, 成田空港, 屋根瓦, 国民年金, 海外留学, 枝垂れ桜, 川下り, ピアノ伴奏, 腹式呼吸, etc. However, when the latter words of these compound words are used alone, they are always pronounced with the original accent. I can’t think of any examples that would fit the scenario you describe. I hope this helps! - Kimi

Walt Poor

Do Forvo and Suzuki Kun seem to implement these complicated rules correctly in every case that you have heard?

Dogen
>>13629155

No, Suzuki-kun often messes up when it comes to compounds. Forvo is usually pretty accurate.

Rafael Ruiz

Thank you endlessly for pitting out this series! I had been studying Japanese for four years then I took a four year break but recently decided to get back into it (even though I'm old and wrinkling now) how does one begin to start changing already made speaking patterns to begin to incorporate the correct phonetics?

Dogen
>>14797087

My pleasure Rafael! It really all depends on the individual, but I've found as little as a month if you record yourself on a regular basis! ^^

Adam Papai

I hope I'll never have to use the word: 自己嫌悪 :D

Dirk Schwartz
>>67828608

I am confused. For one, I don't think your statement is even correct. Both OJAD and Forvo give 男性 as 平板, as does Wadoku. Which brings me to the second point: What do you mean by "falling intonation" ? Is it possible you are mistaking the general very slight falling pitch over ANY expression for an actual drop making this pitch accent 中高 rather than 平板? In that case I encourage you to spend more time on listening exercises because it feels like you're not hearing the pitch right.

PH

Listening on my iPhone; for some reason the lessons (6,8,9)keep stopping and I have to keep restarting; anyone else having trouble? Hopefully just transient.

Dogen
>>19938676

That’s quite odd—never heard of something like that before. Hope it starts working normally again soon!

Nero

So do works like 弁当, which I would assume to be 平板, become 中高 with お？ As in お弁当. Or does it change in some other way?

Dogen
>>27405486

Hi Nero! 弁当 is actually one of the exceptions, in isolation it's 中高 with the downstep on the と, but when it is preceded by お the entire compound becomes 平板. Hope that this helps!

Abhi Ilindra

I just realized something: I'm 90% sure that when All Might yells 緑谷少年！！！！ in the My Hero Academia anime, it follows the same pitch accent pattern as your 白人男性 example.

Dogen
>>28899124

Hi Abhi! I've never seen that anime, but it makes sense that the character would say it that way!

Ole Stegelmann

Wow this lesson was eye-opening. I had never noticed the 中高 pattern in compounds before. But now that I know it, it is like a missing puzzle piece was put into its place, and the picture I had been seeing for quite a while finally makes sense. It is kind of obvious now. Thank you very much for this insight!

Dogen
>>29311533

Hi Ole! Yes, this one is definitely a game changer—once you become aware of it you start hearing it everywhere! Good luck with your future studies!

Kenny Wang

I think this episode + episode 8 were excellent, and showed some real payoff for everything this series has covered up until now. Keep it up, Dogen 先生、よく頑張った！

Andrew Ellison

Hello Dogen, Thank you for your phonetics video series! It seems like in long compounds the accent tends to fall on the first multi-mora syllable in the latter element (long vowels or ～ん syllables). I first saw this idea in a paper that tried to examine intrinsic Japanese phonetic patterns based on how accent patterns were applied to loan-words, and the examples you had (based on coffee) seemed to fit with this pretty well.

Dogen
>>3401518

Really appreciate that Andrew. First mora in the latter element is definitely one of the most common patterns, but 3 from the end is also very common for longer words, like 外国人登録証明書, and so on. That being said, 3 from the end can also be first mora haha. Interesting with regards to the syllable notion, I haven't heard that before, but it certainly does make sense. One of the things that I love about compounds in Japanese is that most of the foreign loan words tend to follow the same patterns as native Japanese words, even when they are mixed together. Thanks again!

Dogen
>>3400324

Thanks Kenny! This is one of the big ones, to be sure. Again, you should immediately begin to hear the 中高 pattern if you weren't aware of it before :) これからもよろしくお願いします！

Axel Hooper

Awesome! Ever since the episodes where you described your resources and the pitch-accent families, I've incorporated flash cards with pitch-accent info into my study routine. I actually unconsciously noticed the 中高 trend but didn't have solid thoughts on the matter, so thanks for clearing it up! Great work!

Dogen
>>3410213

Thank you very much for watching and supporting! Glad to hear that you are making a lot of progress! Also I'm sure that now that you're conscious of the 中高 trend you will be able to use it to your advantage. Good luck with the rest of your studies! 😁

Timothy Chambers

明けましておめでとう！ As always your explanations are very clear and to the point. Also this video helped me clarify some patterns I noticed in listening practice, once the earlier videos helped me figure out what to listen for. One minor quibble/suggestion - for the examples at the end, it would have been nice to have them the first time without the visual pitch overlay, then the second time with. That way we could initially try to "catch" the accented mora just by ear, and then find out the second time if we were right or not. Keep up the great work :D

Dogen
>>3446434

Thank you very much for the comment. I actually debated on this one; visuals first is good for reinforcement/intial exposure, while visuals second is good for application, as you mentioned. That being said, I felt that I was still in the process of introducing a fairly important concept, and thus went with visuals first :) I'm thinking about doing a practice only lesson soon with visuals second :) Thank you for your feedback and happy New Years! 今年もよろしくお願いします！

Julian Bentham

Hi Dogen, this video has reminded me a rule I heard in a Japanese linguistics class, but I'm not sure if I've remembered it properly (or how consistently it applies), so just wanted to run it by you and see if you've come across something like it in your research. Basically the rule I heard is that for most compounds with 5 mora or more, the drop occurs after the 3rd-from-last mora, unless it is a vowel sound, vowel extension, or small tsu, in which case the drop moves to the left and occurs after the 4th-from-last. It seems to apply for the examples given in the video, so just wanted to get your thoughts on it. Btw thank you so much for making such a comprehensive course. My brother's recently started studying Japanese and I've already recommended it to him!

Dogen
>>35755320

Hi Julian! Certainly in many compounds with 5 or mora, the downstep will occur on the 3rd from last mora, but not in all words like this, as you mentioned. For example in キャラメルマキアート the downstep will occur on the third to last mora (ア), but in 金髪美女 (arbitrary example) it will occur on the second to last mora, and there are many other examples like this. In general your rule is very good, but I recommend the advice I give in this video (listen for nakadaka in compounds) instead, as I've found that it's a bit more flexible, but still very useful in guiding learners into getting a feel for how words combine in Japanese. Incidentally, the -3 rule actually goes beyond compounds, and I talk about it at length later on in the series, starting with lesson 59, which can be found here: https://www.patreon.com/posts/japanese-episode-25635423?cid=21422460 Cheers!

Adam アダム

Hey Dogen, I recently thought it'd be a good idea to watch your pitch accent series and boy I'm so glad I did, I can't believe I sounded so bad before! I have a question regarding the 't' sounds in Japanese. When I pronounce mora such as て or た, or any 't' sound in Japanese I was wondering if it is suposed to be a clean 't' sound or a sort of muffled one. The clean sound I reference is creating the 't' with your teeth closed or close to being closed at the start of the sound, like how we'd pronounce 'tea' in english, whereas the muffled one you could create a sort of muffled 't' by just using your tongue by pressing it up against the roof of your mouth or back of the top teeth and retracting it. In my studies I had never thought about it too much but now I'm wondering what the proper way of pronouncing it is. Thanks for the great content!

Dogen
>>36130486

Hi Adam, thanks so much for signing up! I actually cover the T sounds in detail in the below lesson (and contrast them with the English T sounds). Do you mind first watching this lesson? Incidentally, the Japanese T's are usually a bit less aspirated than English T's, and I talk about this phenomenon in lesson 33 (also linked below). Cheers! The T sounds: https://www.patreon.com/posts/japanese-episode-19701427 Aspiration—part of the K sound lesson: https://www.patreon.com/posts/18574535

ソラ

Hi Dogen, I have a question a little unrelated to the video. Should I just get used to the patterns and being more mindful of them, when talking and listening, and not think about memorising individual words' patterns and meaning or should I do something else?

Dogen
>>37160270

Hi Sora. At first I would encourage you to simply try to listen for them in conversation. Once you’re able pick up on each of the four patterns without too much difficulty, I would encourage you to then try to memorize at least the important rules that I’ve laid out in this series, as doing this will enable you to know the pitch-accent pattern for many words without ever needing to look up said words individually. It may also be useful to memorize the pitch-accent for words that you frequently use individually. For example, I often talk about writing in Japanese, so I memorized the pitch-accent of common writing related words. Hope that this helps!

Caleb Cassady

Hi Dogen! Thank you so much for taking the time to make these videos. They're incredibly helpful and I'm 100% certain that my spoken Japanese is going to improve drastically. I had one question about the "5+ mora" rule, as right after watching this episode I came across the word 宗教的行事 in my textbook. Would words like this that use 的 without な still be included in this rule? In the audio recording, it seems to be 中高 (I'm pretty sure I hear the downstep after ぎょ), but I still don't completely trust my own ability to recognize this! Thanks.

Dogen
>>3861145

Hello Caleb! Thank you for the support, really appreciate it.

Alexandre Dubray

Hi Dogen! First of all, thanks for the very instructional videos. I hope you and your family are well! To get to the point, this is my first time learning a language in a deep way. My mother tongue is French (I must say, for the record, that I am from Belgium though) and I never took the time to learn correct English pronunciation/phonetic. In the case of pitch accent, I've no awareness of how it actually works. I mean that I do not know how to actually change my pitch when speaking. I can get a feeling of what to do by "hmmm"-ing but it is really difficult to actually change it when trying to apply the different patterns. So in the end, when trying to repeat after you, in the videos, it gives something very unnatural and also wrong. Moreover, my ears are not trained (yet :-)) to listen such pattern so I cannot tell by myself if I pronounce correctly and I have to rely on a third-party program (I saw in one of your comments somewhere that you mentioned PRAAT so I use that). Do you have actual tips/training routines to help me overcome this difficulty? Or maybe it will just come with time (I guess that hearing the change in patterns will come with time:-) ). Thanks in advance!

Dogen
>>38762299

Hi Alexandre! I think to a large degree this will come with time—for about the first two weeks after learning about pitch-accent I couldn't distinguish between some of the patterns, and was thus unable to reproduce the patterns. That said, after a few weeks of continuing to listen to native speech while also looking at graphics I was able to learn to hear the differences. Things may seem daunting at first, but stick to it, and certainly listen to as much native speech as possible. If you don't notice significant progress after a couple weeks then it may be good to try and supplement things with a program such as PRAAT to analyze native recordings vs. your own speech—try without it at first though!

kusterdu

Hi Dogen, At the beginning of the video you mentioned a link to the Shinmeikai dictionary, but I can't seem find it.

bodelu

Dogen, Is the resource you are reading from in this video the Suski "Phonetics of Japanese Language" book that you've mentioned, or some other resource? If it's another resource, which one? Thanks!

Dogen
>>4322858

Brett, That is one of the sources, but I actually don't recommend that source for learning Japanese phonetics, as, despite the title, most of the book isn't actually dedicated to phonetics. What are you trying to learn? Maybe I can offer you some more specific advice based on this :)

Sonic Swain

Thanks Dogen, the native audio is wonderful. I think it's a nice observation that after a decent amount of input, you can "feel" most of these 5+ mora words. I'm no newcomer to Japanese but pitch-accent is relatively new to me (I fell into the 男性 頭高 trap too). Yet I was able to correctly say every example at the end before hearing it. I think rules are less important for this group of words because the 'feeling' is more intuitive. Thanks again for a great lesson!

Dogen
>>44241062

My pleasure, Michael! Glad to hear that you're getting a lot out of the series, and that you thought that the 'feeling' observation was useful—I've personally always used this and it's worked well for me. Cheers!

Mackenzie

Hi Dogen, I was wondering if I could get some advice... I've spent ten years one and off (mostly on the last 6 years) learning Japanese. Because i haven't been able to really practice speaking a lot (I live in rural Canada lol) I instead have had more focus on the reading/writing aspect. I'd like my listening and speaking to improve, and discovering this series late last year was huge for me because in my ten years of learning (and learning about learning) japanese, NO ONE has ever mentioned pitch accent!! I figured I should stop dead in my tracks and learn pitch accent, but while doing this series, I realized just how limited my vocab is (for clarification I'm like N4 level vocab/reading, and N5 for listening/speaking skills). After that realization, I became a little lost as to how best to proceed. I checked the comments section and found someone with a similar problem of vocab, and you'd recommend they stop with pitch accent for the moment and focused on acquiring a better handle on vocab. I thought I'd do the same, but then ran into the problem of... do I learn the vocab with their pitch accent? And how would be best to do that? I don't want to learn all the vocab with the wrong PA and end up having to re-learn them and make things more difficult for me down the line. Your videos with MattvsJapan have eased my fear of not sounding "native" (it's okay to not be perfect, and I no longer will expect that of myself), but I want to make as much progress as possible after spending so many years of learning... so I want to do it right this time. So my question is: how would you recommend I proceed? When it comes to learning more vocab, is there a method that works best so that I learn the PA for those words properly (flashcards, etc)? Thanks so much for this series, Dogen, and for taking the time to respond to people and help them! It's been a big help to me on my journey to attaining the best Japanese level I can.

Dogen
>>49521641

Hi Mackenzie! I think that for your situation specifically it would probably be best to watch through episodes 33, and also episode 59, but to do so rather quickly and not worry too much about memorizing every rule. You really just want to concentrate on the important rules, like the 4 mora 2 kanji rule, the compound noun rule, the -3 rule in episode 59 etc. That is to say, there is a lot of information in these lessons, but by watching them and memorizing the big ones (I usually say 'this is a big one' haha), as well as just being exposed to the other ones, and not necessarily memorizing them, you'll still make a lot of progress when it comes to your pitch accent. I assume that based on your current position it's not necessary to memorize the pitch of most individual words. If you instead memorize the important rules then you can immediately know the pitch accent for hundreds of words, and save yourself a lot of time. So again, go through the lessons I mentioned above, try to memorize the important rules, and after you're done going through the lessons (this shouldn't take more than two months) then I would go back to studying Japanese holistically while also getting a lot of native input. Then, once you feel like you vocal and grammar skills are a bit higher, you can come back to the series and try to memorize some of the less important rules, or just familiarize yourself with the rules you initially skimmed over. This is probably the best approach—again, the most important thing will always be getting a lot of native input, so always try to do a lot of that. Hope this helps! Cheers.

switz mka

Hallo Dogen! I noticed the word 自動車 at the end, which i had previously seen as both 中高[2] and 平板. Is there a dominant pronounciation to use in words like these, or is it just an exception?

Dogen
>>51572838

Hi Swat. In words in which there are multiple pronunciations listed then it's my understanding that it's OK to pronounce said word either way, and I believe that I talk about this is lesson 31, the pitch-accent FAQ lesson (though it could be earlier). Cheers!

malthe

Never thought I'd have to say バニラクリームフラペチーノ to learn Japanese...

Mark

Hi Dogen! I’m currently working through Japanese: The Spoken Language, the introductory textbook series by Eleanor Harz Jorden. It’s all in romaji and from the eighties, but I really like the accompanying audio, grammar explanations and the fact that pitch accent is indicated for everything. Anyway, as for my question: in the first lesson of the textbook, for example, the pitch accent of きのうつくりました is indicated as LHHHHHHLL (going up on の and down on し). This seems to be consistent with the way it’s pronounced in the audio accompaniment. This confuses me a bit, because I assumed that the pitch accent would be LHL LHHHLL (with both words retaining their ‘isolated’ accents), as both are nakadaka words. When I put the phrase into the OJAD prosody tutor, it indeed indicates LHL LHHHLL as the correct pitch pattern. Do you have any thoughts on this phenomenon? Thank you so much for your unparalleled phonetics course!

Dogen
>>53188242

Hi Mark. Great question—in this circumstance, 昨日 is being used as an adverb, which makes it 平板. This is a rare phenomenon (when certain words are used as an adverbs the pitch-pattern changes to 平板) that only affects a small handful of words, and I talk about it a little bit later on in the series. Hope this helps, cheers!

Mark

That makes a lot of sense, I hadn’t thought of that! Thank you for your quick answer and clarification. Are you referring to episode 73 on adverbial counters? I haven’t yet gone through the entire series but that was the only thing I saw that pertained to pitch accent in adverbs.

Dogen
>>53266199

Hi Mark! Yes it is, although there are some other words, again such as 昨日, which follow this trend as well, though I don't cover them in my series, as I don't yet know all of them. Cheers!

Landon Drewes

Hi Dogen! Thank you for the wonderful series. I majored in Japanese in college and unfortunately only had two classes that stressed pitch accent, so your videos have been very useful. Is the resource you are reading from in this video a Natsuko Tsujimura book? Just curious. Thanks again!

Dogen
>>62688445

Hi Landon! Glad to hear that you're enjoying the series thus far! I'm not Reading from Natsuko Tsujimura here, but rather a handout that I directly received from my teacher at Keio University. Thank you very much for your support!

Some_guy

Hi, again. I noticed something interesting as I was doing some phonetics practice. The word もう is normally 頭高, but when used in もう少し, the whole phrase seems to become 中高. A similar thing happens with そう, which is also 頭高 and likewise the phrase そう思う seems to become 中高 as well. Are these changes following similar rules as found in Lesson 9? I might just be hearing things incorrectly, but I thought I'd ask to be sure.

Dogen
>>64132570

Hi, hardy_hermit_crab! Thank you for the comment! Yes! You have made a great discovery, and what you’re hearing is absolutely right!! (^^) But grammatically, it’s a bit different from the compound words that are mentioned in this lesson. もう is adverb, and basically pronounced 頭高. But this もう in もう少し is a special case, and it modifies another adverb 少し. In the case like this, もう is pronounced 平板, thus もう少し is pronounced as in 「moO-S(U)KOshi」with devoiced す. As for そう思う, そう has different meanings depending on the context, and when it’s used as an adverb like “そう”思う, it’s pronounced 平板 as in 「soO」. Thus そう思う is pronounced as 「soO-OMOu」. The site I put the link below has many example sentences using そう. If you would like to, please take a look. そう in the last two examples are pronounced 平板. https://www.nihongomaster.com/dictionary/entry/601/sou But in my opinion, you don’t have to memorize those grammars I mentioned above, you just keep making discoveries like you did this time, and practice pronouncing them, so that they’ll come naturally to you! Good luck with your studying! -Kimi

Brodie Port

On NHK digital pitch accent dictionary and forvo, 男性 is pronounced with a falling intonation. (dann\sei). I often find conflicting pronunciations between your videos and NHK/ forvo which confuses me and I don't know which ones correct. In this video, when 男性 is pronounced fastly at 4:14 it even has a falling pitch going against what is notated.

TheStellarJay

Hey Dogen, I struggle very hard with hearing fluctuations in pitch (up or down) on the characters: あ、え、い、お、う、ん Do you have any tips for these ones in particular? Thanks.