Goals and Commitment to fluency through practice (6-month plan)

Hi Herb! This is a great thread. I’m just curious how often were you studying the 5000 words in the first month? Was it about 1 hour like you do with Glossika now and then with your other reading/activities on the side or was it hours and hours of flashcards at first to build that word bank in your head? Did you start from a pretty beginner level and how long did it take for you to feel comfortable with it? I ask these questions because I was building my own word frequency cards (based on my experience of needs and what I hear or see in Brazil) for a while and cramming them for hours a day but then I’d eventually burn out after about 3 weeks and quit everything haha. I’ve done that more times than I can count over the past year. But if I can visualize an end to the misery than I can get back on the horse.

Also with Glossika, I keep reading you don’t like the new online version but I can’t find an offline version of Brazil Portuguese to purchase where I am so I’m a bit stuck with the online version. Anyway, is your main gripe that you can’t use the mp3s on your bike and only on a laptop? Or is there something really different from the book version to the website version? I enjoy being able to walk up and down stairs while listening to Pimsleur so if I can’t do that with Glossika then that’s disappointing. Since my wife is Brazilian I considered just recording her saying the sentences for me if that’s the case and putting them into flashcards to loop the audio 5-10x.

Thanks for the support.

I might have gotten up above 1 hour per day, everyday, at the worst of the 5000 words. It was mostly about an hour per day for 6-8 weeks

Early on I was releasing about 200 news words per day (can’t remember excactly) and I knew about 1000 non-cognates starting out.

(About 1 year of fairly recent but poor high school French worth becase we are false beginners. I took 4 years of French 55 years ago and I’m would be mad about how poor it was – if there was any point – knowing what I know now. They taught us so much that was just wrong and taught us so little through doing it badly.)

If you know SRS (Anki) systems you’ll realize that releasing 200 cards per day can easily create days with many times that number of reviews due: At the worst, for about 10 days or so it was up around 800 cards per day which I did throughout the day, mostly in dead time.

It was about a month to get through the whole deck 5000 / 200 = 25 days but the first week or so the release rate was down around 100 before I increased it to get the words loaded into my head quickly.

Once the first pass was through I could pretty much read with dictionary support, first my own arrangement of dictionaries (Primarily Linguee, GoogleTranslate, Wiktionaraire, and a list of almost 3000 non-obvious cognates I bought that included weak mnemonics but primarily a lot of etymology which helped with memorization.)

It took about 2 months to get the majority of it to “mature” – which means the cards are only reviewed something like once per month or longer. It’s been on maintenance now for while at 99% plus and it runs bout 60-80 cards per day.

You can read some of the others posts to see what dates (from late Dec start) my Anki deck reached various levels of maturity.) Some of it might be over in the original post that is here on LingQ somewhere. After a bit I realized this “Personal Accountability” was expected in this Challenges section of the forum.

My theory that is different from most of the experts seem to recommend is to “Load the first 5000” word into my brain as rapidly as possible/practical so that free reading of “comprensible input” becomes possible with a variety of intersting material.

It seems obvous to me this is a recommendation that is sorely underrated.

The sooner you can read most of the material you enjoy the sooner the massive comprehensible input can “take over” the main learning load.

The goal is not to primarily “learn” the words - thus why I think of this as brain “Loading”, but rather to get them recognizable so that reading them in context becomes clear enough to be self-sustaining.

Send me a private email and I can probably help you find the old Glossika stuff. HerbMartin at that Google mail place.

To me, the “new” Glossika isn’t even the same program even if it uses (some of?) the same sentences. The old audio program from 2016 is rapid fire, there is no stopping unless you stop the playback manually.

You get hit with new prompts in English/native language almost faster than you can say and listen to the correct answer for about 25 minutes straight per “lesson” which is called a “Day”.

I was really excited at the prospect of using the web version and prepared to buy at year at about $100+ until I tried it.

It was far more than just not being easy to put on the phone and do in the car or while biking. It didn’t even feel like the same program but rather like “just another language site”.

I do 2 per day, occasionally sneaking in an extra one, with a regrettable lapse of 10 days once, and there are 312 “days” (lessons) in the full 3 levels.

Last night I finished Lesson 20 of Level 2, or 124 of 312 with 188 to go…

Doing 1 per day will take more than 10 months even if there are no lapses. That’s too long for me to commit – doing 2 is harder but once I am on the bike with the audio running it’s easier to keep going than to do “another day” 6 months from now.

It’s not a sure bet I’ll finish, but it seems more likely than not at this point.

I seriously doubt that even 1% of the people that bought this program would ever finish all 312 – my guess would put it at more likely 1 in a 1000 or less.

This is probably the reason the program had to be changed so radically when it went to the web and a subscription basis. You can’t expect people to keep paying if they stop doing the program after a month or 2 and in the current environment you can’t sell a year’s membership to everyone due to competition from 100’s of other language programs.

The program on the web is fundamentally different. As to recording your wife that is pretty good as a substitute, but very time consuming 1000 x 3 items, and you really need SOME form of pseudo-random prompts and responses.

If you decide to do this look in “JoyTan” (I am a member of his Slack group) and he knows how to produce long audio visual language programs with some automation support. We’ve been talking.

My guess is you would never get it done as a “hobby” level activity and it would eat up a lot of time to produce.

If I didn’t have Glossika I would probably be using Pimsleur, Michel Thomas or similar. I like “Learn in Your Car” best but there isn’t enough of it available. Only about 100 or so short lessons (5 minutes?)

To be very clear, getting started and staying going with the Glossika audio program is very difficult for me – but it works.

LingQ is easy by comparison even at the “insane level”.

Anki is easy by comparison, even at 400+ reviews per day.

Doing the 5000 words vocabulairy is essential in my opinion and doing the reading (whether you use LingQ or not) is also essential.

Doing some form of audio program is also essential to develop speaking and fluidity…

I don’t do “grammar” as such – at all – and no consider that “grammar is evil” and actually harms our progress if you “learn” any grammar beyond you ability to immediately use it.

Grammar is for me a “clean-up” progress – which is true in our school systems also. No child learns grammar before being able to speak, read, and even write fairly fluently.

When you “know” grammar it just slows your speaking to a stop and interferes actively with any attempts to write simply.

I now treat “grammar” as a review occasionally or as “vocabulary” and currently I am review all of the major tenses with BrainScape decks (another flashcard system) but it takes me only an hour or 2 to “load” an entire tense/mood conjugation set into my head for most of the common regular and irregular verbs because I mostly know them or have familiarity alread.

I also worked out a “patten” for French verbs which gives me quick mnemonics for most French verbs. This only works for about 90% of 90% of all verb conjugations but like “5000 words” is covers 99% of what I’ll use in real production and lets me read easily.

BTW, Unlike a lot of common advice, I consider making the majority of your own review decks or audio programs to be a poor use of time – I do add to and edit my lists or build special purpose versions.

It is far more efficient to make use of existing decks – at least for 80-90% of your work) than to try to build the perfect deck.

I buy some word lists and grab all sorts of free lists – though as a programmer it is common for me to take an existing “list” that isn’t set up for flashcards and convert it, or to take the words from (sets of) books by dumping them to text and just “uniquing and sorting” them.

Even then my project to load “all the locutions” from Wiktionaire hasn’t completed. I keep putting it aside because I’m too busy learning to make more decks.

BrainScape is not a “great” system – Anki is far better. But as many free decks as Anki has, BrainScape probably has 100 TIMES more and they are pretty high quaility though no multimedia as far as I know.

Last, if you look elsewhere in these threads you’ll see that I use mnemonics and etymology extensively to make it quicker and more efficient to do brain loading of the vocabulary.

(I’ll reply to this post since I can’t reply to your reply haha.)

Thanks for the details!

1 Hour is more than doable and not going to burn me out. I’ll get back to it and use a minutes restriction on this and add more minutes after completing my other daily life To Do list. Honestly my life spins a bit out of control if I don’t limit myself on language study because I just get too obsessed with cramming but then I turn the rest of my life into a mess and then I just burn out eventually.

I appreciate flashcard study although i really hate Anki. I use flashcards deluxe or Quizlet. Anki has always given me migraines for some reason. FD and Qz do what I need as far as SRS, review games, and adding computer generated Audio from Google or Amazon voice.

Speaking of which as far as Glossika I suppose I could just download Amazon’s BR-Pr voice into my flashcard app (very simple to do and can be randomized) but I’ll email you to see about finding those proper Mp3s for much higher quality audio and join the JoyTan slack to see the benefits of that. But I’m really impressed how far computer generated audio has come which is likely do to Amazon Alexa and Google Home being in everyone’s homes these days.

I’m using a google doc daily diary for my Personal accountability which my wife checks in on and gets on me if I haven’t updated it. But I should probably move it onto this site to keep my LingQ use up as well.

I agree that making decks is a huge time waster because I can get too obsessed with always looking for more cards to make and focus less on actually practicing the cards. I should also note that this is a major reason why I hate Anki haha. The few times I’ve tried to be an Anki user it always took me hours to set up cards. Flashcards Deluxe (my main app) and Quizlet (I just use this for extra review tools) are so freaking easy to set up cards super quick but I understand Anki offers tons of extra options that users demand and better stats to have a better clue of your progress but ya, huge time waster for me hence why I dropped it.

Anyway I do agree finding pre made cards is very beneficial.

For listening I’ve been searching desperately for Podcasts, Youtube channels, TV Shows with accurate transcriptions/subtitles which has been a really shocking process because there’s not a huge source of native Brazilian Portuguese content with (non auto generated) subtitles/transcripts. I also wanted to watch Dubbed TV shows I like on Netflix but the netflix transcripts are so inaccurate compared to the dubbed audio. It’s really baffling how bad it is, I don’t know how Netflix doesn’t require the subtitles to match the dubbed audio. But recently I’ve found enough of the above and discovered Amazon Kindle and Audacity for Brazil has extensive amount of material so I’m good to go.

I don’t know " flashcards deluxe or Quizlet" so I’ll check them out and perhaps wish that I hadn’t spent money on BrainScape.

When I say “Anki” it generally means “any suitable SRS flashcard program” unless I’m obviously comparing more than one. My original one for Arabic in 2002 was “Pauker” which was actually pretty close to modern Anki. For the time it was very advanced and free, open source too.

JoyTan is not organized to sell you anything but he has free open source tools and produces a LOT of YouTube “audio flashcard” content because he can do it efficiently. (Many languages.) And he’s (or they) is really nice.

He also has an Android app that offers to pay a small amount for people who will pronounce their native language. I donate my time whenever I do that.

Working some type of diary is really helpful so whatever works for you. I learn a lot (about myself and my own techniques) simply from asking questions that I’ve never considered deeply or in detail.

BTW, it’s not “cramming” if you do it every day. :slight_smile: :slight_smile: :slight_smile:

I can understand the obsession and actually use that rather than simply motivation. Every human trait and behavior has some context in which it is valuable and useful (and the opposite of course.)

Strategic obsession and constructive procrastination both have their proper place and role to play. (I am serious about this.)

I really don’t see the problem with suitable material and find the “accurate transcipt” location to be a very small problem. Any transcript is usually good enough and I actually prefer YouTube auto-generated transcripts and wish all videos had them. (Most do if they are in a single language.)

I wish a lot of the good “French” teachers would stop speaking English since that turns off the auto-translation from YouTube.

Also, I can take a target language transcript and throw it into Google Translate (5000 characters at a time which is only mildly annoying or I’d write a script to to it) or I can pull it into LingQ and just read it with assistance.

This lack of understanding your problem means I am likely missing some key detail that is troubling you so please feel free to explain and I’ll try to see if I’m solving it some way or know a workable method to mitigate the issue.

I actually like the mismatched subtitles, either in target language or my English, because it keeps me on track but allows me to still listen for the target language without having “the answer” handed to me.

It’s fun to “hear” the REAL DIALOGUE and catch them offering weird or even incorrect translations or transcription.

Also, I frequently extract my own NetFlix transscripts in order to run entire seasons of a series (or several series) though my vocabulary extractor. Then we can learn the words we don’t know and hear (or even read) better.

The main keys for me are:

  • Effective methods (this is my specialty)
  • Consistently doing all of the work every day
  • Staying with it when life intervenes
  • Having fun

2 or 3 out of 4 will work but having all 4 is best.

Quick note:

Optimism remains and even increases. The simple subvocalization method works and becomes automatic and less of a burden the more I practice it.

My guess is the behavior will largely automate and pass to unconscrious, especially as skills improve. Then it will only be necessary to concentrat on using it for speech that is more difficult to hear and understand.

Reason for this update:

I went back to David’s ( https://fluentlistener.com ) content after skipping a few days playing with subvocal shadowing and my comprehension on that material had improve a lot.

The understandable parts were easier and the hard parts were about 50% understandable.

KEY TAKE AWAY:

His methods are STILL VERY IMPORTANT. Subvocal Shadowing doesn’t replace them, at least not totally, but it make it easier for focus on the truly difficult parts and to consolidate the things you learn from working those parts.

Moral: Don’t throw babies away with bathwater.

For non-Native English speakders: “Don’t throw the baby away with the bathwater” is a well-known English idiomatic maxim that means, “Don’t lose the good parts when removing the waste or the unnecessary elements of a method or solution.”

(reply #3, again I can’t reply directly)
Flashcards Deluxe is cheap which is good but only works on mobile devices. The developer responds to all emails and forum posts if you have a question. But I love that app. I think the website has a card search feature too you can source from.

Quizlet is on both phones and web portal and a good source of cards (but I don’t know the card quality) but I use it only for the test features. Not my main driver since it starts asking for subscription fee for some features.

Ya I suppose transcripts/subtitles for TV shows / Movies that aren’t 100% accurate don’t bother me too much because generally actors are speaking very clearly. But for podcasts there aren’t any apps or sites yet that make transcripts automatically (apparently Google Podcasts is working on this but it’s not an available feature yet) so I strictly listen to ones that blessed the world with providing transcripts because I really can’t handle podcasts for native speakers yet without some type of reading material to follow along. Especially because of how native speakers blend their words or just completely drop the first part or last part of a word. I don’t remember this being an issue for Spanish but Brazilians drop so many sounds I feel like it’s as hard as learning English pronunciation but the difference you can’t know English pronunciation just by reading it where technically you should know how to speak a Portuguese word according to proper phonetic pronunciation but so many words are not often fully spoken and sometimes they skip right over saying certain words but if you check the transcription it’s supposedly there. I remember the first time I tried to listen to one podcast for native speakers and transcribe it myself and it sounded like complete gibberish. After I read the transcription I was shocked because I knew all the words but they said these words NOTHING like you’ll find in Portuguese audio lessons for language learners haha. Depending on the channel YouTube struggles a lot with this as well so it’s almost like learning a different language sometimes. Thus I really really REALLY value transcripts to study how the language ACTUALLY sounds.

I’ll go ahead and Post It Note the methods on my desk to keep motivation up. Effective methods certainly appear to be your specialty so I’ll be following your posts to see whenever you find new methods to try out.

Though I don’t know about Portuguese it would seem likely that it wouldn’t be a lot worse than French for having good content available with closed captioning or transcripts.

A couple of suggestions:

Try to find “TED Talks” from a Portuguese version – they exist for French, and likely for Portuguese including the Brazilians.

Try to find “course work” for school children. This was initially harder to find for French but once I rigged my searches to look in the right places it was a gold mine.

Google trick that few people use: Specify the site and give only a top level domain

  • site:PT | site:BR

As to hearing that is precisely what my new method (Subvocal Shadowing) and David Tolman’s methods are designed to work explicitly.

My evaluation is that these methods work extremely well, and my only reservation is that no one else seems to have describe any of this in any detail though I’ll bet lots of people use the methods either unconsciously or without explaining their method.

One thing about making methods explicit is that they are easier to improve and refine or to use explicitly where they work the best.

My ability to hear has gone up markedly in the last week since finding David’s work and even more in the last 3 days since starting to use “SubVocal Shadowing”.

Combining the 2 is something that I’ve only started to test tonight and it seems to work well also – perhaps even better.

Thanks for the David Toleman/subvocalizing tip. I checked out his site and listened to his Echoing demonstration. Then it dawned on me how I’ve echoed before to really get the grasp of different English accents (Irish/British/Australian/Other American accents). I was watching an Irish TV series and I ended up echoing the main character so much I (no joke) got kind of stuck in an Irish accent for a few minutes and I was even thinking in an Irish accent. It freaked me the hell out hahaha. I also remember this happening when I was around a guy from New Jersey with a thick accent and I kept copying all this guy’s expressions for 10 straight days I came back to Texas and it took me a day to snap myself out of doing a Jersey accent. As far as Portuguese, I remember a day a few months ago when strangely everything I was hearing I was understanding. The same way you describe listening to that news broadcast. I wasn’t echoing outloud but it occured to me I might have been subvocalizing. I think that’s possible because I was on a marathon of cramming vocabulary and listening and listening and doing more listening and echoing then. Then one day it was like a switch flipped and I was suddenly understanding the radio, what people on the street were talking about, what the taxi driver was saying. It was freaky as hell because before that day I just thought I was beginner level. So ya I think echoing/subvocalizing are really important methods. And one more thing, when I was a kind memorizing scripts for plays at school, I wasn’t echoing but I was reading out loud again and again and again and again. It was probably this that created the feedback loop and I was able to memorize lead parts in plays. So yes, fully support these methods.

I love you post.

The post’s with encouragement and congratulations are helpful but posts like yours where someone descrives their own experience of using things similar (or instead of) that work for them are even better since it gives me ideas.

David is primarily emphasizing working on sections repeatedl with echoing and call/response or echoing aloud while the speaker continues for more advanced people.

To me these are not the same technique and they do address two overlapping but different aspects and skills.

The repetition on one difficult phrase or passage is mainly about hearing (with learning to speak that way as a secondary effect.)

The continuous (and concurrent) echoing (some call it shadowing but I think that is different also) is more about hearing the full flow of the speech.

Then my technique of SubVocal Shadowing is almost totally focused on hearing to understand in the moment.

It doesn’t help improve future hearing by teaching the surprising ways that things can sound that we think we already know, but it’s really about just “getting it” right now.

That’s the way I separate them – and to be explicit I use all 3 for different types of material each day.

One might expect that Subvocal Shadowing is my favorite because it is my own discovery but the main reasons that it’s my most common technique are:

  1. It’s immediate – it works to better understand speech right now.
  2. It can be done without any special equipment and without tieing up our eyes and hands (while working, driving, excercising, lying down to sleep etc.).
  3. It can be alternated with passive listening – checking in whenever the thread is lost.
  4. It helps me so much when listening to news where the speakers will continue long enough to “check in” but where the speakers also change frequently.

Working a passage really helps me to hear difficult things that were literally beyond my current level of audio decoding.

This is almost magical too – I’ve made notes of phrases and passages (mostly from David’s material) that were literal “unhearable” and then a few minutes, hours, or even days later “it’s impossible NOT to hear it.”

This is such a big effect that without the notes even I wouldn’t believe it later.

The main difficulty is that it focuses so much time and effort on such short passages that we only learn a little, at least consciously.

It’s difficult to know how much that will affect our future hearing ability on other passages with similar types of deletions, liassons, de-emphasis, speed, etc. that native speakers use constantly, and that other native speakers hear without effort most of the time.

For a while, I really worked at shadowing/echoing aloud but it so gets in the way of my hearing what the speaker says that it doesn’t seem to be the most efficient.

I don’t know the proper mix and I am reluctant to stop the two aloud practices in case they are critically important to rapid progress.

Glossika and Rosetta Stone along with more vocabulary work also help but it’s now important for me not to “coast” through that vocubulary work but rather keep pushing.

For several days I’ve been pushing Glossika to do an extra lesson or even 2 so now it’s past Level 2, Day 30 – it seems impratical to keep it up at this level for another 59 days but I like the effects.

Mainly I wish that I was talking to real people more and that there was a better way to fine tune my pronunciation though I work on that as constantly as possible.

I’m thinking to start a “daily recording” session where I just speak for 30 (?) minutes without allowing myself to leave French. Sort of a “private podcast” (because I doubt anyone else would want to hear it.)

Though making it public would motivate & even force more improvement probably.

I’ve started using your subvocalizing method while listening to podcasts, Herb. What a brilliant discovery.

I’ve looked for something similar to David Tolman’s method for languages other than French but not really found anything comparable. speechling.com has a dictation feature (you listen to a sentence in your target language then try to type out what you heard). Do you think that would have some of the same benefits of David’s method?

That is wonderful news, please keep it up – and please let me know when it works and even if it doesn’t. (That’s how you can best thank me unless you can come up with additional or even better methods.)

As to the “listen-write” method that is basically a Dictée or very similar and mathilda thought that was a good idea and I do something like that for at least learning to hear and read so yes, I think it is worth trying at a minimum.

As to it being brillian, it does make me feel pretty smart but it really is so simple that it’s hard to believe it isn’t common knowledge.

2020-04-28 (almost 4 months into the 6 month plan)
48,000 known words
18,000 Lings (created manually)
79th insane streak day
143 (of 312) Glossika lessons
Rossetta Stone: Finished Unit 1 of Level 4 (only 7 more Units remain)
1000+ words into the English to French 5000 word list.

I have really pushed on Glossika and Rosetta Stone recently, including 6 days of upping the count to 3 lessons per day – and still being able to keep up. Where once 1000 new words per day was pretty easy, finding 100 or so per day can be a challenge (though I don’t focus on this and I am down to being largely satisfied with 100+ links to continue my streak but mainly to read for listening better.)

At the rate of 3 lessons per day, it would be possible to finish all 312 Glossika lessons but that is not something I’ve decided to do, or even necessarily required each night. Even 2 lessons are already a lot of work.

It is getting difficult to find “new words” in LingQ through ordindary reading that interets me, even by the measure of LingQ which counts every inflected form.

Listening mostly to Deuil Interdit (Bosch by Michael Connelly) and reading it in LingQ
(Also still listening to Sapiens, Fluent French, and several all French podcasts, and news etc.)

We’re watching “Osmosis” (not very good but watchable) and various other NetFlix titles in French with French subtitles, or sometimes with English subtitles…

(When alone, I am trying to turn off the subtitles more and more.)

A caution on the method of Subvocal Shadowing needs to be offered to avoid disappointment:

After an initial boost the rate of improvement doesn’t seem to continue – my thinking on this is that it just allows you to understand (more of) what you already could read but it doesn’t necessarily create new improvements at that same higher rate.

This still leads to additional improvement, since we learn by increading our “comprehensible input” this allows us to increase our rate of comphrehesion more quickly and to more quickly lock onto a new voice or new material.

It’s probably the case that what we “can hear” is closer to what we could read in a single pass – without resort to going back and figuring out the sentence and piecing together the meaning.

Listening is different from reading because we can’t control the speed – it just keeps coming and we must either understand it on the first pass or we mostly miss it.

It’s harder to learn most new words through listening since we can’t resort to a dictionary quickly enough to be of much use for most words.

For me, this means that reading the same material, either before or after listening, it an important step for learning to hear as well as for hearing more in the moment.

This all remains something to be developed and fully understood…

With only 63 days left on my 6 month plan, fluent listening seems less than likely – though still vaguely possible.

Rather than becoming discouraged by this likely truth or giving up, I am continuing to push to find out what really is possible for me to accomplish.

One encouraging thought:

I have just about as many days to use LingQ as I have used so far, and around half of both Rossetta Stone and Glossika remaining along with enough time to complete the “brain load” of the 5000 words from English to French to improve ACTIVE vocabulary.

Given what I have learned in the past 4 months, and that much of my current methods were begun only about 2 month ago, this means I have between 50% and 100% chance to improve.

That is not nothing even if the clock seems to be rushing on…

Great update.

Since I found your post it’s really gotten me to focus on self studying better than I ever have and follow your learning plan (not doing as many variety of activities every day but definitely spending at least 3 hours on 2-3 activities) giving me much more structure but also feeling much more relaxed because I’ve had a breakthrough in figuring how to to search for youtube videos with closed captions, using the Duolingo stories for Portuguese has been really helpful/entertaining, and I’ve changed how I use flashcards (less forced memorization and just more reviewing sentences I’ve vacuumed up from lessons I import into LingQ to get my brain used to seeing grammar structures or remind myself of new vocabulary). I’ve had my Lingq account for 5 months but barely touched it much but I’m sticking to using it for everything I do now. Thanks for the inspiration.

It sounds like you are in that level of your language learning that Steve talks about where you have achieved a great level of vocabulary but then you feel like you’re still struggling a lot which begins the long trek through the desert not knowing how far you have to go but in reality you’re making progress even if it doesn’t look like or sound like it (in your case with your listening skills). Anyway, I hope you’ll be able to at least see the edge of the desert by the time you get to 6 months. How are the speaking lessons going with the tutors?

It is truly gratifying that my posts have had even some small benefit to you and it sounds like you have found methods that work for you.

As to the tutors that really only helps me with speaking, mostly through my choice, and my largest concern is not doing that enough due to various scheduling issues (his and mine) as well a procrastination, or perhaps reticence, on my part.

Yesterday I upped the release rate (new cards) on the French to English deck to 65 which will get me through the 5000 deck by July 1. 65 may not sound like a lot of cards to those who don’t do spaced repetition but already the daily reviews are up around 170 and that will grow, probably to around 300=400 in the next couple of weeks

It reached over 800 for a bit when doing the English to French direction at 100 new cards per day. It’s easy to get overwhelmed when new cards and returning cards start adding up. though I am also just suspending, i.e., retiring, cards which are well known to me.

Maybe a “trek through a desert” is a good metaphor since the main issue is not to know how to push forward, in which direction, if there is somewhere reachable on the other side, if one is wasting energy/time/resources traveling the wrong route, or even going in circles.

I can force vocabulary and reading, even speaking, but forcing “hearing” is much more difficult to arrange.

My latest idea (just last night) was to revisit InnerFrench (Hugo) and similar who are speaking strictly French but as teachers who are careful with both vocabulary and pronunciation, and who speak a bit slowly.

During the first 5 (starting at the beginning some 100 or more podcasts or so back, it’s almost easy for me to hear and understand virtually every word. It only requires a bit of attention to keep up and I even have a moment or two to figure out a word here and there while still hearing what he is saying.

So the main question will be if this leads progressively and steadily to better understanding or is merely an indulgence which feels good but accomplishes very little.

We learn best when we are fastest and most succesively when we are mostly succeeding but working right at the edge of our capability. (Deep Practice aka Deliberate Practice).

62 days remain in the plan, and a sense of urgency helps with learning a great deal.

Posting this because it’s useful perhaps for someone else to hear and as a record, and to say thank you:

Today was not a great day, French was Ok, but work, life, health, stuff. If it weren’t for the plan, this thread, and you folks I would probably have blown off LingQ, Rosetta Stone, and Glossika.

I am way tired.

LingQ is done (barely at 105) and I’ll somehow do the other two though minimums probably.

So thanks everyone who has contributed or who even reads this.

Thank you.

Keep it up - your commitment to learning is very impressive.

Because I’ve been doing 3 Glossika “Days” (lessons) per day a big milestone just fell:

  • Glossika Level 2, Lesson 52 is complete. (156 of 312 lessons in all 3 Levels)

…but there is good news and bad news in that:

  • Glossika is 50 complete
  • There are 156 Glossika lessons remaining to complete all the rest of Level 2 and all of Level 3.

Which is the good news and which is the bad news depends on the way we view this.

As a realist, for me it’s simply that half is done and half remains.

Perhaps the best way to look at it is to realize that were I not improving greatly there is no way that I would have persisted with this every day and there is really no way I would be willing to accelerate this to 3 lessons (1.5 hours) per day.

So this means that I have the opportunitey to double the improvement Glossika has already provided.

It sounds pretty good that way.

Aren’t we all looking for a way to “double my learning”.

Also, work on my new method to improve speaking ability has begun in earnest – not much accomplished yet but at least now the work is actually being done and it’s no longer just a planning.

Hearing is improving but the daily improvement is still maddenly slow and irregular.

Subvocal shadowing is probably the single most value tool base on effort versus efficacy that I have so far found.

It’s making the difference between “hearing some words” and actually “understanding” each new levels of difficulty.

Newscasters are in and out of reach for me – but only a week or so ago they were almost always beyond my ability.

It’s sort of like a radio station that does in and our as we drive cross country (maybe you need to be old to really understand this metaphor so perhap it’s like the when the network bandwidth gets crowded or cleared.)

There are several “internal voices” that are likely to interfere with understanding difficult speech passages so I’ve given them (temporary) names:

  1. The Translator – the voice trying to put it into our native language
  2. The Day Dreamer – the voice when we feel disconnect and bored. “Gee I wonder what’s for dinner…”
  3. The Doubter – “Is that really what she said?”, “Will this every work.?”

Just silencing these voices is worth the effort to shadowing just to avoid the interference within our own minds.

That the subvocalization actually allows us to match again the external sound and thus by activating our motor neurons we engage our mirror neurons involved in understanding other people and increase our understanding much closer to our current theoretical maximum.

And so far, I can find no significant disadadvantages to using it. One must remember to do it, and put in the effort but that is actually pretty trivial and can be done without any extra equipment or even with our eyes closed.

One fun test of Subvocal Shadowing is to put on something like a TedX talk in French and then simply do something else but every so often just “pop up” realizing I haven’t got a clue but almost immediately “check in” by begining to use the subvocalization.

I will remind you – and myself – the most important think beyond USING the method is REMEMBERING that “ambiguity” is acceptable.

Relax and stay with the shawod of the speaker…

LingQ 90th insane streak day – I love LingQ, have learn a lot but there really have been no breakthroughs
Rosetta Stone Unit 14 of 20 is complete.
48,069 words known, and 19,204 LingQs

Glossika Level 2 Lesson 63 is done – need to pick up the speed to finish by July 1. (167 of 312 complete)
Anki 5000 French to English – 20 is mature/retired, 66% unseen (turn passive vocab to active)
Brainscape – not doing this enough or regularly enough (time)
Reading – Bosch/Connnelly and Sapiens books are getting close to finished but I still can’t follow the audio unless I’ve already read the text.

My new “pronunciation improvement” trick is the new podcast quality mic I bought which has a jack for my headset which gives me immediate feedback on my pronunciation even when not recording. Helps immensively to hear my mistakes. See thread here for details:

Also the Facebook group “Pronunciation Best Practices” (language agnostic) is well-worth checking out and the PDF “Kjellin-Practise-Pronunciation-w-Audacity.pdf” by Olle Kjellin in “Files” section is definitiely worth the time to read (and study) 22 pages.

Hearing & Understanding fast spoken French remains a problem though sometimes the news is almost understandable if I really concentrate.

Discord French group remains a disappointment. Try it, but I am not going to waste very much time there.

My “words known” now go up only a fews words per day despite always having more then 100 LingQs. The material I am reading simply has no new vocabulary.

In more than a week since breaking 48,000 words I’ve added less than 69 despite 1200 more LingQs.

Clearly you’re tackling this goal with a lot of commendable dedication, and I have to applaud you for that, but I do think your overall methodology might create some imbalanced results that are hard to qualify on the short run. I’m speaking specifically about the 48,000 known words. I’m not sure what your criteria is for marking those, but based on my experience I’m skeptical about learning that many words in a 5 month period. All the more so because your overall words read is about 800K which, for 5 months starting from scratch, sounds very high to me for conscientious reading, and the same time sounds very low to have learned 48K words from it. (As a comparison, I think I tend to be a fairly liberal marker of known words, and over the course of 3 languages, I’ve been in the ballpark of about 20-25K known words marked for every 1,000,000 words read.)

To be clear, in the long run, this won’t matter, because if you stick with it, those 48K words will solidify just fine, but on the short run you’ll have this imbalance that is represented by your stats of known words vs. experiences with listening comprehension.

I’m just bringing this up so that expectations can be compared, and to point out that most students should expect to have to read more words over a longer period of time, and hit different comprehension benchmarks when using LingQ.

Thanks, all attempts to help are appreciated.

You analysis is however faulty so it’s leading to incorrect conclusions in this case.

There is no really need to mark known words so those are words I didn’t mark as unknown which is pretty easy to do: You either know them in context or not and LingQ’s idea of a known word includes all of the inflected forms which means about a 3 to 1 inflation factor.

On the other hand, I did mark 19k LingQs manually because I turned off the automation for that and go to a lot of trouble to find unknown words (in the blue) by primarily to check and mark a lot of phrases.

Doing 100 to 900 LingQs per day (213 average) adds up to a lot of careful reading.

In fact I am more concerned about reading too slowly and wasting too much time with LingQs and such.

Being a VERY good reader, in general, it is really know issue to read such a SLOW pace.

My reading comprehsion is very near fluent.

French as you know has a lot of cognates so reading in context you will occasionally have to check one for alternate meanings but that’s pretty quick with LingQ to help.

Thanks and feel free to offerr any other thoughts or suggestions…