Datasets:
Tasks:
Automatic Speech Recognition
Languages:
Swedish
fix more markdown; convert to UTF-8
Browse files
README.md
CHANGED
@@ -1,4 +1,4 @@
|
|
1 |
-
THE WAXHOLM CORPUS
|
2 |
|
3 |
The Waxholm corpus was collected in 1993 - 1994 at the department of
|
4 |
Speech, Hearing and Music (TMH), KTH. It is described in several
|
@@ -8,9 +8,10 @@ information on the Waxholm project can be found on the web page
|
|
8 |
http://www.speech.kth.se/waxholm/waxholm2.html
|
9 |
|
10 |
|
11 |
-
FILE INFORMATION
|
12 |
|
13 |
-
SAMPLED FILES
|
|
|
14 |
of the speaker is coded by the two digits after 'fp20' in the file
|
15 |
name. The smp file format was developed by TMH. Recording information
|
16 |
is stored in a header as a 1024 byte text string. The speech signal in
|
@@ -32,7 +33,7 @@ range=-12303,11168 ; amplitude range
|
|
32 |
```
|
33 |
|
34 |
|
35 |
-
LABEL FILES
|
36 |
Normally, each sample file has a label file. This has been
|
37 |
produced in four steps. The first step was to manually enter the
|
38 |
orthographic text by listening. From this text a sequence of phonemes
|
@@ -105,14 +106,15 @@ retroflex stops were not labelled as retroflex in the original .mix
|
|
105 |
files ( d, t after 2D and 2T have been changed to 2d and 2t).
|
106 |
|
107 |
|
108 |
-
REFERENCES
|
109 |
-
|
110 |
-
B., Gustafson, J., Hunnicutt, S.,
|
111 |
-
L., Nord, L., de Serpa-Leitao, A., and
|
112 |
dialogue data collected in the WAXHOLM project" STL-QPSR 1/1995,
|
113 |
KTH/TMH, Stockholm. Bertenstam, J., Blomberg, M., Carlson, R.,
|
114 |
-
|
115 |
-
|
|
|
116 |
N. (1995). The Waxholm application data-base. In Pardo, J.M. (Ed.),
|
117 |
Proceedings Eurospeech 1995 (pp. 833-836). Madrid.
|
118 |
|
|
|
1 |
+
# THE WAXHOLM CORPUS
|
2 |
|
3 |
The Waxholm corpus was collected in 1993 - 1994 at the department of
|
4 |
Speech, Hearing and Music (TMH), KTH. It is described in several
|
|
|
8 |
http://www.speech.kth.se/waxholm/waxholm2.html
|
9 |
|
10 |
|
11 |
+
## FILE INFORMATION
|
12 |
|
13 |
+
### SAMPLED FILES
|
14 |
+
The .smp files contain the speech signal. The identity
|
15 |
of the speaker is coded by the two digits after 'fp20' in the file
|
16 |
name. The smp file format was developed by TMH. Recording information
|
17 |
is stored in a header as a 1024 byte text string. The speech signal in
|
|
|
33 |
```
|
34 |
|
35 |
|
36 |
+
### LABEL FILES
|
37 |
Normally, each sample file has a label file. This has been
|
38 |
produced in four steps. The first step was to manually enter the
|
39 |
orthographic text by listening. From this text a sequence of phonemes
|
|
|
106 |
files ( d, t after 2D and 2T have been changed to 2d and 2t).
|
107 |
|
108 |
|
109 |
+
## REFERENCES
|
110 |
+
Bertenstam, J., Blomberg, M., Carlson, R., Elenius, K., Granström,
|
111 |
+
B., Gustafson, J., Hunnicutt, S., Högberg, J., Lindell, R., Neovius,
|
112 |
+
L., Nord, L., de Serpa-Leitao, A., and Ström, N.,(1995). "Spoken
|
113 |
dialogue data collected in the WAXHOLM project" STL-QPSR 1/1995,
|
114 |
KTH/TMH, Stockholm. Bertenstam, J., Blomberg, M., Carlson, R.,
|
115 |
+
|
116 |
+
Elenius, K., Granström, B., Gustafson, J., Hunnicutt, S., Högberg, J.,
|
117 |
+
Lindell, R., Neovius, L., de Serpa-Leitao, A., Nord, L., & Ström,
|
118 |
N. (1995). The Waxholm application data-base. In Pardo, J.M. (Ed.),
|
119 |
Proceedings Eurospeech 1995 (pp. 833-836). Madrid.
|
120 |
|