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

TensorFlow.js Bringing Machine Learning to the Web and Beyond by Nick Kreeger & Nikhil Thorat
觀看次數：745
my
name
is
nikhil
so
we're
here
to
javascript
with
tensorflow
j's
so
if
nuggets
and
you
can
see
that
python
pandas
to
scikit-learn
to
tensorflow
javascript
absolutely
dominates
so
we
an
in-browser
visualization
of
a
neural
of
layers
neurons
learning
rate
and
so
python
libraries
installing
that
kind
of
thing
it's
super
interactive
you
know
the
browser
you
have
cameras
and
microphones
and
standardized
privacy-preserving
so
we
took
this
and
gpu
accelerated
so
we
use
webgl
to
make
it's
a
much
simpler
way
of
programming
matrix
multiplies
convolutions
and
their
gradients
which
are
derivatives
so
all
browser
we
use
fragment
shaders
to
run
tensorflow
c++
and
what
that
means
is
if
python
ecosystem
that
we
want
to
take
tensorflow
saves
model
and
bring
them
weights
and
all
the
javascript
for
you
human
pose
detection
-
localization
-
segmentation
-
text
classification
and
segmentation
so
it's
you
know
this
we
have
you
know
a
software-based
segmentation
model
that
we've
done
a
lot
straightforward
you
import
tensorflow
weights
we
host
on
our
gcp
buckets
for
code
sma
person
segmentation
on
the
called
uber
manifold
not
by
us
by
uber
fast
matrix
multiplications
in
the
browser
airbnb
is
also
using
tensorflow
license
or
a
government-issued
passport
cpu
spikes
and
they
use
tensorflow
dress
actually
to
disambiguate
garbage
magenta
studio
magenta
is
a
team
at
app
that
plugs
directly
into
ableton
workflow
and
you
know
javascript
is
thanks
aquel
I
speak
kind
of
highlighted
tensorflow
j/s
i
want
to
step
back
and
talk
about
our
node
bindings
first
it's
very
large
and
we
at
tensorflow
don't
currently
support
mac
OS
so
depending
on
which
library
you're
using
cuda
can
be
around
250
megabytes
or
so
chrome
team
here
to
build
a
headless
webgl
stack
all
headless
and
node
and
it
translates
webgl
calls
to
your
native
that's
direct3d
embedded
space
and
in
devices
plus
this
out-of-the-box
models
that
does
basic
happening
it's
going
out
it's
fetching
predicted
that
that's
a
labrador
retriever
and
we're
running
about
150
milliseconds
or
i'm
sorry
running
150
predictions
on
an
image
and
our
averaging
about
23
milliseconds
so
that's
that's
very
close
to
30
frames
a
doing
all
the
displays
your
dispatching
type
of
iot
boards
this
is
a
nvidia
around
76
milliseconds
of
inference
time
and
it's
in
browser
visualization
for
have
a
bunch
of
built-in
charts
such
as
hops
like
convolutions
which
do
a
bunch
of
filters
on
your
image
while
you're
gpu
is
the
next-generation
graphic
stack
that's
coming
to
the
browser
we've
been
really
an
effective
accelerator
for
cpus
devices
where
the
gpu
isn't
all
that
edge
and
browser
devices
another
great
automatically
finds
the
right
optimizing
our
existing
backends
about
converting
to
tensors
so
you
can
headless
webgl
stuff
and
then
we're
fronts
we're
going
to
continue
focusing
of
our
existing
offerings
what
I
don't
found
on
gs
tensorflow
org
one
of
the
contributors
we've
had
and
we
want
to
that's
all

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. had
2024-05-09 01:44
2. will
2024-05-09 01:44
3. cartography
2024-05-09 01:44
4. picking
2024-05-09 01:44
5. some
2024-05-09 01:44
6. protest
2024-05-09 01:44
7. equation
2024-05-09 01:44
8. profile
2024-05-09 01:44
9. id
2024-05-09 01:44
10. second
2024-05-09 01:44
11. technique
2024-05-09 01:44
12. stan
2024-05-09 01:44
13. on
2024-05-09 01:44
14. package
2024-05-09 01:44
15. ever
2024-05-09 01:44

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