1
00:00:00,000 --> 00:00:05,200
Practice and review pronunciation. The
pronunciation assessment uses Azure Speech

2
00:00:05,280 --> 00:00:10,480
for United States English. It is feedback
on a recording, not a level exam.

3
00:00:11,040 --> 00:00:16,240
Step 1. Choose a phrase. In Practice
pronunciation, choose the suggested

4
00:00:16,320 --> 00:00:20,480
phrase or Another phrase and type exactly
what you want to say.

5
00:00:20,560 --> 00:00:22,600
Listen to the example if you need to.

6
00:00:22,880 --> 00:00:26,160
Practice a short phrase before moving on to
a longer one.

7
00:00:26,480 --> 00:00:31,840
Step 2. Record a take. Press Record me, say
the phrase, and stop the

8
00:00:31,920 --> 00:00:36,400
capture. Hear my take allows you to check
that your voice was heard.

9
00:00:36,480 --> 00:00:40,960
If there is noise, silence, or cut-off
words, repeat the recording before

10
00:00:41,040 --> 00:00:45,280
evaluating it. Step 3. Request the
evaluation.

11
00:00:45,520 --> 00:00:50,640
Choose Evaluate and compare the recognized
text with what you intended to say.

12
00:00:50,840 --> 00:00:55,760
Depending on the mode and available data,
pronunciation, fluency, intonation, and

13
00:00:55,840 --> 00:01:00,400
completeness will appear. If it could not
be evaluated, do not interpret the

14
00:01:00,560 --> 00:01:05,520
problem as a zero grade. Step 4. Use the
feedback to improve.

15
00:01:05,920 --> 00:01:09,760
Repeat the words or fragments that need
practice and compare attempts.

16
00:01:10,160 --> 00:01:13,040
Completeness makes sense when there is a
reference phrase.

17
00:01:13,520 --> 00:01:18,480
The filmed example uses a synthetic voice
to test the process; its score does not

18
00:01:18,600 --> 00:01:22,320
measure a person nor represent an expected
result for everyone.
