系統設計者的話：學英文根本不用背，你講中文，你有在背中文嗎？學英文也是一樣的道理，那是自然而然就會的東西，用背的方式最後肯定都是不行的。 更多...

最好的學習就是有英文的父母，不斷的跟你說話、耐心的教導你，時間長了，你也就慢慢就融入其中了。

但是我們沒有英文父母怎麼辦？看影片學英文就學的起來嗎？學不起來。

我們從小到大，也看過了不少英文電影，至少也看過1、200部也跑不掉，但是為什麼看了這麼多的英文電影，英文還是很爛呢？

主要的原因是我們沒有從基礎性的強化學習。舉個例子，西方的幼兒出生時，他至少聽了各種大人講英文至少聽了2年，才慢慢開始說英文、與大人互動。

而且都是一字一句慢慢的說、慢慢的教、漸進式的、慢慢的、越來越習慣、越來越聽的懂、越來越會說的過程。

那麼我們看的影片、電影過程中，都是大人成熟型的講話速度、方式、語調，我們怎麼可能跟的上？

一旦跟不上，思想中就會產生覺的好難哦、我不行種種念頭，那麼產生了這種潛在式的排斥心理作用，那麼你就更難學的會了。


所以我們必須打好基礎、深厚的基礎，萬丈高樓平地起，唯有打下深厚的基礎，後面你走的路才會越來越順。

程式界裏有句話，程式債欠的越多，最後就只能放棄。學英文也是一樣，英文單字債欠的越多，你就直接上來文法、句型、語調，最後也只能放棄。

基於上述的思維，我打造了這套系統，從最基礎的單字學習起，不斷的強化、加深你對這個單字的聽力，透過各式各樣的影片、各式各樣的人、各種不同的場景。

就只是為了訓練一個單字的聽力，無數的人跟你講同一個單字，講個20遍、50篇、100遍(且不重覆)，我就不信學不起來。


這套系統做出來後，我自己在使用的過程中，確實真的能聽出來他們講的這個單字，而且他們講的速度再快也能聽的出來。

能聽的出來，就知道他講的大概意思；能夠知道他講的大概意思，就能夠用幾個簡單的單字回應；能夠回應，也就能夠對話；當能夠對話，就能夠再修正文法、句型、語調。


我的目地就是希望自己至少能聽的懂人家在說什麼，用英文跟人家對話，大家共同努力、學習。若有其它想法或對這套系統設計的建議，也可以與我聯絡：ichich2013@gmail.com

Learning From Machines by Ashi Krishnan | JSConf.Asia 2019
觀看次數：390
intersection
of
computational
of
our
cognition
sometimes
in
unsettling
michaela
pareo
nieto
and
he
has
chromatic
hatches
the
crystalline
farm
banana
a
toaster
it
performs
this
convolutional
filters
now
convolutional
filters
are
basically
photoshop
filters
each
neuron
in
a
convolutional
layer
has
and
each
convolutional
layer
applies
a
kernel
a
kernel
is
a
matrix
of
numbers
each
pixel
in
a
neurons
receptive
field
labelled
image
that's
something
where
we
know
what's
in
it
it
outputs
predictions
and
then
we
math
these
visuals
inverts
that
so
this
visualization
is
recursive
to
compute
the
spiders
will
become
corgis
and
the
teleporter
accident
of
our
neural
networks
than
this
one
I
think
the
lizard
juxtaposition
has
retina
your
retina
our
retinas
they're
there's
a
layer
of
ganglions
which
are
the
photoreceptors
the
rods
and
cones
photoreceptor
that
photoreceptor
is
backwards
and
there's
a
hole
in
the
photoreceptors
and
there's
10
times
your
ganglions
than
that
so
it
can't
be
a
one-to-one
mapping
and
to
the
bandwidth
of
our
optic
nerve
is
10
megabits
which
cameras
are
about
100
megapixels
it's
receptive
field
that's
a
hundred
or
so
photoreceptors
that
is
that
are
divided
region
so
when
there's
no
light
on
any
of
them
the
ganglion
doesn't
fire
when
ganglion
fires
very
weakly
when
only
the
they're
just
ganglions
that
naturally
image
coming
in
from
our
photoreceptors
a
hundred
times
while
retaining
vitally
it's
gonna
hit
the
optic
chiasma
where
cortex
so
there's
some
slice
of
tissue
our
visual
cortex
has
a
receptive
field
like
gradients
and
surfaces
and
objects
convolutional
neural
networks
like
loops
these
pyramidal
neurons
that
become
consciously
aware
of
it
improving
and
these
loopback
pyramidal
cells
in
serotonin
receptors
different
kinds
of
pyramidal
cells
respond
to
serotonin
familiar
with
serotonin
in
its
starring
reuptake
inhibitors
effects
may
occur
most
serotonin
in
your
serotonin
receptors
and
their
binding
produces
different
effects
so
5-htt
1a
receptors
tend
to
be
inhibitory
somewhat
drowsiness
inducing
5-htt
3
receptors
in
the
brain
they're
associated
sensations
anti-nausea
drugs
are
frequently
5ht
3
antagonists
there
is
another
serotonin
receptor
one
that
the
pyramidal
cells
in
this
is
the
5-htt
2a
receptor
this
is
dissolves
releasing
molecules
of
fly
surgit
acid
diethylamide
into
your
gut
now
LSD
doesn't
bind
to
5
ht
3
receptors
has
no
trouble
acrossing
the
blood-brain
finds
a
pyramidal
5-ht
to
a
receptor
and
receptor
trapping
LSD
inside
which
would
you're
looking
at
may
be
nourishing
the
like
proto
qualia
which
will
be
toaster
or
say
don't
these
skiers
look
delusions
I
mean
that
sticker
does
look
there's
the
head
there's
the
body
and
if
stateful
it's
a
continuously
refined
groundwork
for
capsule
networks
sera
sabor
nicholas
frost
and
geoffrey
hinton
fixed
multi-layer
neural
network
like
a
architectures
recurrent
convolutional
neural
networks
that
ruminate
upon
images
making
better
classifications
or
there
are
adversarial
examples
for
the
convolutional
neural
networks
with
classifier
it's
vacillating
and
haw
and
reconsider
our
own
perceptions
at
many
levels
that
gives
our
perceptual
paradoxically
being
able
to
second-guess
continuously
reconsidering
and
refining
shifting
hypotheses
about
the
state
of
we're
tripping
face
or
well
asleep
dreamt
up
by
a
generative
adversarial
the
creator's
forgeries
from
true
faces
forgery
it's
doing
well
if
it's
forgeries
are
detected
we
back
propagate
the
discriminator
I
changed
the
names
our
movements
ruminate
on
a
space
of
you
may
have
seen
in
long-forgotten
see
a
face
there's
a
bunch
of
neurons
in
vector
embedding
some
mapping
from
faces
vector
here
it's
not
totally
without
neurocognitive
so
we've
hooked
an
electrode
up
to
a
recommend
spinning
around
this
detaches
you're
perceived
body
the
one
with
12
of
position
has
a
neurocognitive
called
an
efference
when
an
efference
is
of
efference
copies
is
as
reflections
we
the
periphery
and
about
10
milliseconds
latency
or
high
bandwidth
our
nervous
coordinated
movements
our
brain
has
to
in
sync
this
attenuation
applies
even
to
the
cerebellum
the
cerebellum
is
very
coordination
like
this
people
with
cerebellar
damage
it's
our
movements
become
jerky
and
laggy
it's
theorized
that
the
cerebellum
body's
able
to
estimate
the
body's
continuously
varying
control
signal
and
once
you've
evolved
it
the
thing
like
both
transformations
from
the
space
of
predictor
encounters
a
problem
when
there
is
an
irreconcilable
discontinuity
they're
both
visceral
and
guttural
they
they're
both
shattering
the
illuminations
of
discontinuities
again
water
smoothing
the
edges
of

相關影片

加Line好友 每日8AM 推播1單字

加入好友

其他人正在學習的單字

1. only
2024-05-04 02:39
2. tentacle
2024-05-04 02:39
3. panda
2024-05-04 02:39
4. music
2024-05-04 02:39
5. specific
2024-05-04 02:39
6. list
2024-05-04 02:39
7. shutdown
2024-05-04 02:39
8. records
2024-05-04 02:39
9. giant
2024-05-04 02:39
10. organization
2024-05-04 02:39
11. insist
2024-05-04 02:39
12. surface
2024-05-04 02:39
13. activities
2024-05-04 02:39
14. joseph
2024-05-04 02:39
15. went
2024-05-04 02:39

本月觀看影片單字排行榜

本月查閱單字排行榜

本月聆聽單字排行榜