InterstellarCG commited on
Commit
af72568
·
verified ·
1 Parent(s): 4be0f0f

Publish DeepSahel Kikuyu Language Model KenLM

Browse files
Files changed (7) hide show
  1. .gitattributes +5 -0
  2. README.md +29 -0
  3. corpus.txt +3 -0
  4. kik_3gram.arpa +3 -0
  5. kik_3gram.binary +3 -0
  6. kik_5gram.arpa +3 -0
  7. kik_5gram.binary +3 -0
.gitattributes CHANGED
@@ -33,3 +33,8 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ corpus.txt filter=lfs diff=lfs merge=lfs -text
37
+ kik_3gram.arpa filter=lfs diff=lfs merge=lfs -text
38
+ kik_3gram.binary filter=lfs diff=lfs merge=lfs -text
39
+ kik_5gram.arpa filter=lfs diff=lfs merge=lfs -text
40
+ kik_5gram.binary filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - kik
4
+ license: apache-2.0
5
+ tags:
6
+ - kenlm
7
+ - kikuyu
8
+ - gikuyu
9
+ - language-model
10
+ - deepsahel
11
+ pipeline_tag: text-generation
12
+ ---
13
+
14
+ # DeepSahel Kikuyu Language Model KenLM
15
+
16
+ A Kikuyu KenLM language model resource for speech and text pipeline experiments.
17
+
18
+ This model is part of DeepSahel open African-language AI infrastructure work, supporting reusable model-development, evaluation, and documentation workflows for low-resource African languages.
19
+
20
+ ## Intended Use
21
+
22
+ - African-language speech and text research
23
+ - Kikuyu model evaluation and benchmarking
24
+ - Public-interest language technology experiments
25
+ - Downstream prototyping for education, health communication, agriculture, and civic information workflows
26
+
27
+ ## Limitations
28
+
29
+ Performance can vary across domains, dialects, speaker conditions, recording quality, and code-switched usage. Human review by fluent speakers is recommended before production use.
corpus.txt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:267684c399f734ca249b21b45b2fdfcc4dea7e591c3c459a03bf7dad8b9100ca
3
+ size 22910078
kik_3gram.arpa ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:036292cd08d79509a4fb3d8946af5ac0c1310e81813c3e19ae647255b07ccec5
3
+ size 132279649
kik_3gram.binary ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0d5be07dc328864b4088567bf6ce6fb40d0f978321b7b5cbbf1227543e0b884a
3
+ size 78490658
kik_5gram.arpa ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:56b9698f1f9c62d6bb2de343237960a91b653393b0e775ed400373944f85663e
3
+ size 425737681
kik_5gram.binary ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fcee58f2b9922b3ab466794563f973f3c9ef73da74e6e014de1d144d7eaa6678
3
+ size 206001890