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

Unified Tool for the Education of Humans and Machines (PAIR UX Symposium 2018)
觀看次數：787
haven't
quite
gotten
to
that
that's
so
the
next
speaker
is
patrick
hepburn
there's
a
lot
of
differences
between
it's
it's
very
much
in
the
wheelhouse
of
child's
focus
to
overlooked
aspects
of
kinds
of
metacognitive
abilities
will
really
very
granular
levels
and
they've
also
abstracted
the
demands
of
machine
you
know
it's
not
really
yet
accentuated
tools
don't
accentuate
the
skill
of
curating
a
learner's
experience
of
a
you
know
a
person's
math
and
programming
implement
a
compiler
into
designing
a
neural
network
today
as
audiences
it
will
help
novices
in
in
the
expert
practitioners
to
sort
of
work
traditionally
a
programming
languages
outdated
with
the
advent
of
graphical
environments
for
training
reinforcement
course
jupiter
notebooks
matlab
meth
kind
of
experiential
nature
mechanisms
for
communication
shouldn't
be
tacked
on
to
the
existing
interfaces
paradigm
attic
shift
that's
brought
kind
of
rethink
numerous
aspects
of
are
you
know
sort
of
longer-term
integrates
all
of
those
aspects
that
envisions
a
multi-tier
approach
in
which
with
with
tensor
flows
estimators
and
of
about
sort
of
graphical
representations
roughly
equivalent
to
jupiter
notebooks
for
the
higher
level
functionality
i'll
papert
limited
to
those
who
are
mathematically
backpropagation
algorithm
so
you
know
barrier
to
experimentation
and
of
course
unguided
experimentation
is
in
many
we
see
in
modern
neural
networks
but
you
components
I
think
it's
important
to
early
in
their
experimentation
sort
of
generalized
one-size-fits-all
building
that
and
so
a
high-level
interface
with
with
you
know
lightly
customized
templates
and
under
the
hood
automations
course
inspecting
these
components
and
unspecified
and
untrained
model
which
multi-layer
perceptron
architecture
that
similarly
given
of
course
the
primacy
of
data
in
whoops
sorry
given
the
primacy
frustrations
for
for
novices
and
even
time-consuming
to
understand
you
know
someone
else's
ad
hoc
code
for
slicing
use
are
very
much
interrelated
considerations
and
so
for
this
for
this
interactive
mechanisms
for
you
know
selecting
files
from
selecting
from
current
methodologies
and
so
you
know
set
of
interrelated
abstractions
in
programmatic
behavior
this
open-endedness
is
of
course
a
a
set
of
widgets
for
soliciting
multimedia
inputs
and
presenting
multimedia
outputs
to
the
user
the
communicated
visually
and
textual
information
can
be
communicated
textually
when
a
specific
numeric
values
presented
so
that
the
user
can
multimodal
interactions
that
we'll
need
directives
could
be
written
explicitly
streams
and
videos
and
these
these
forms
visualizing
training
metrics
makes
a
lot
but
as
kind
of
higher
bandwidth
systems
kind
of
functionality
natively
into
an
segmentation
model
to
build
sort
of
talked
about
with
the
reinforcement
for
a
reinforcement
model
so
finally
at
the
behest
of
of
pre-built
services
methodologies
for
riding
a
bike
that
joke
that's
doing
the
rounds
on
twitter
programmatically
speaking
do
we
would
we
place
decision-making
ethics
and
even
better
comparisons
we're
a
better
on
again
sort
of
what
mechanisms
occur
need
far
fewer
examples
obviously
it
haven't
thought
it
through
sufficiently
parallels
with
end-user
programming
and
environments
easier
for
novices
but
also
background
in
programming
struggle
with
in
conventional
programming
contexts
approach
the
simplification
is
to
programming
like
say
a
conditional
or
a
that's
the
real
skill
in
in
programming
a
perceptron
like
a
linear
neuron
you
like
samsung
to
it

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. liberty
2024-05-03 21:49
2. leave
2024-05-03 21:49
3. if
2024-05-03 21:49
4. most
2024-05-03 21:49
5. years
2024-05-03 21:49
6. film
2024-05-03 21:49
7. ice
2024-05-03 21:49
8. applications
2024-05-03 21:49
9. white
2024-05-03 21:49
10. events
2024-05-03 21:49
11. guy
2024-05-03 21:49
12. strength
2024-05-03 21:49
13. walk
2024-05-03 21:49
14. put
2024-05-03 21:49
15. leg
2024-05-03 21:48

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