Tools
Sorted by:
Try your query at:
 |
 |
 |
 |
 |
 |
Results 1 - 10
of
12,359
Table 3: Statistics of word lengths in the LDC dictionary and in the learned dictionary.
2004
"... In PAGE 4: ...3 Word Length Statistics In this section, we present statistics of the word lengths in the LDC dictionary as well as in the self-learned dictionary extracted from the align- ment. Table3 shows the statistics of the word lengths in the LDC dictionary as well as in the learned dictionary. For example, there are 2 368 words consisting of a single character in learned dictionary and 2 511 words in the LDC dictionary.... In PAGE 7: ... This should result in an improved dictionary. An alternative way is to use the word length statistics in Table3 as a prior dis- tribution. In this case, long words would get a penalty, because their prior probability is low.... ..."
Cited by 3
Table 2: Recognition results using Dictionary Learning
1996
"... In PAGE 3: ....2. Experiments In our n0crst set of experiments we carried out all the steps described in the previous section, with exception of retrain- ing. Table2 summarizes the n0crst results and their compar- ison with the baseline system that does not use alternative pronunciations. In experimentA1we generated alternative pronunciations which do not result in homophones in the dic- tionary.... ..."
Cited by 21
Table 4: Results Dictionary Learning (GSST)
"... In PAGE 4: ....3. Results We have implemented this system for bi-directional transla- tion between English, German and Spanish in our scheduling task. Table4 shows the performance of parser and subse- quent generator on transcribed data. Evaluation of the system based on speech decoded by the JANUS-2 recognizer is still underway.... ..."
Table 4: Results Dictionary Learning (GSST)
"... In PAGE 4: ....3. Results We have implemented this system for bi-directional transla- tion between English, German and Spanish in our scheduling task. Table4 shows the performance of parser and subse- quent generator on transcribed data. Evaluation of the system based on speech decoded by the JANUS-2 recognizer is still underway.... ..."
Table 4: Results Dictionary Learning (GSST)
"... In PAGE 4: ....3. Results We have implemented this system for bi-directional transla- tion between English, German and Spanish in our scheduling task. Table4 shows the performance of parser and subse- quent generator on transcribed data. Evaluation of the system based on speech decoded by the JANUS-2 recognizer is still underway.... ..."
Table 3: The top 50 words in the learned dictionary and real dictionary of WSJ88 database learned (100 docs) learned (1000 docs) learned (3000 docs) real
1998
"... In PAGE 9: ..., which are well suggestive of the overall subject of the database. Table3 gives the top 50 words in the learned dictionary when 100, 1000, and 3000 documents are examined, and the top 50 words in the real dictionary of WSJ88 database. It shows that as more documents are examined, the learned dictionary converges to be more close to the real dictionary.... ..."
Cited by 1
Table 3.1: Performance after one lifetime (appr. 500 language games) of an agent with a fixed dictionary learning behaviour and a non-linguistic agent
Table 3.2: Performance after one lifetime (appr. 1000 language games) of an agent with a fixed dictionary learning behaviour and a non-linguistic agent
Results 1 - 10
of
12,359
