Dataset Viewer
Auto-converted to Parquet Duplicate
audio
audioduration (s)
1
10.8
utterance_id
stringlengths
11
11
tier
stringclasses
1 value
language
stringclasses
1 value
split
stringclasses
1 value
source_type
stringclasses
1 value
source_file
stringclasses
180 values
source_url
stringclasses
0 values
article_id
stringclasses
0 values
license
stringclasses
1 value
license_basis
stringclasses
1 value
duration_s
float64
1
10.8
snr_db
float64
10
33.8
speech_fraction
float64
0.6
1
span_start_s
float64
3.56
2.07k
span_end_s
float64
4.99
2.08k
text
stringlengths
2
232
text_source
stringclasses
1 value
transcript_grade
stringclasses
0 values
lid_top
stringclasses
62 values
lid_top_prob
float64
0.09
1
lid_p_en
float64
0
0.41
lid_p_target
float64
0
0.78
lid_source
stringclasses
1 value
text_flag
stringclasses
4 values
text_lid
stringclasses
53 values
text_lid_prob
float64
0.03
1
text_lid_match
bool
2 classes
text_script_match
bool
2 classes
text_repeated_trigram
bool
1 class
text_dup_count
int64
1
12
text_chars_per_s
float64
1.14
215
amh_u008658
1a_lessons
amh
train
FSI
FSI - Amharic Basic Course - Volume 1 - Unit 08.mp3
null
null
PD-US
17 U.S.C. § 105
1.25
21.6
0.904
225.46
226.71
ስስ ሰሰ ሳሳ ሲሲ ርር ሴምሚልራማሁለመሎህሆላሜሩሕሊሠረሮሡሬሪሓሖሗሉሱሒሽሙሀሢሞሟሣሯሦሤሥሃሴ ም ማ መ ለ ሚ ሮ ረ ሪ ራ ል ላ ሜ ሠ ሖ ሩ ሁ ህ ሣ ሙ ሡ ሢ ሃ ሎ ሀ ሉ ሥ ሞ ሆ ሻ ሗ ሂሻሼሂ ሽ ሓ ሟ ሕ ሒ ሬ ሑ ሱ ሊ ሦ ሼ ሄሹሏሄ ሏ ሌ ሇ ሤ ሐሺ ሯ ሔሷሔ ሷ ሺሑሇሌሐ ሹ ሸሧ ሾሾ ሧሿሸ ሿ ሶሶ �, �
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.37817
0.02614
0.00003
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.7072
false
true
false
1
156.8
amh_u000375
1a_lessons
amh
train
FSI
13c.mp3
null
null
PD-US
17 U.S.C. § 105
1.92
14
1
481.75
483.67
ስስ ሰሰ ሲሲ ሴርሳሩምሡሚልመሽራማሎሱህሴ ር ም ማ ሚ መ ሡ ረሮሯሁሜሆላለሕሠረ ል ሮ ለ ሙሹሪሖሗሊሓሒሢሄሉሷሟሃሣሙ ሁ ሠ ሜ ሩ ሢ ሖ ሪ ራ ሗ ሞ ሑ ህ ላ ሣ ሳ ሟ ሃ ሆ ሂ ሻሻ ሱ ሎ ሓ ሒ ሽ ሉ ሀ ሥ ሷ ሕ ሇ ሼ ሄ ሹ ሊ ሬ ሦ ሺ ሯ ሤ ሏ ሌ ሔሂሇሀሏሥሞሑሦሌሔ ሸ ሐሺሤሬሧሐ ሾሾ ሧ ሿሿ ሶሼሸሶ �, �
openai/whisper-large-v3 (pseudo-label, language set)
null
cs
0.96582
0.00191
0
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.6629
false
true
false
1
102.08
amh_u007157
1a_lessons
amh
train
FSI
56b.mp3
null
null
PD-US
17 U.S.C. § 105
1.18
14
1
264.61
265.79
ስስ ሰሰ ሲሲ ሳሳ ርር ሴምሚልመሽሡህማለሁራሎላሜሬሠሕሊሮሩሆሀሱሖረሗሃሢሄሉሴ ም ማ መ ሚ ረ ሮ ለ ሠ ሙሪሂሻሓሣሟሙ ል ሡ ሪ ሜ ሩ ሢ ላ ሁ ህ ሖ ሃ ሉ ሀ ሆ ሗ ሎ ሑ ሞ ራ ሕ ሟ ሥ ሣ ሓ ሻ ሬ ሂ ሱ ሇ ሊ ሒ ሽ ሄ ሏ ሌ ሷ ሯ ሦ ሼ ሹ ሔሷሒሥሏሇሹሔ ሺሑሞሯሦሐሌሼሸሤሺ ሸ ሤ ሐ ሧ ሾ ሿሧሿ ሶሾሶ �, �
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.76953
0.00709
0.00044
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.6334
false
true
false
1
166.1
amh_u003825
1a_lessons
amh
train
FSI
34b.mp3
null
null
PD-US
17 U.S.C. § 105
1.46
16.1
0.815
291.08
292.54
ḍḨᵍᵁᵋᵈḍ ḏḠḏ Ḡ ḥḘḉᵉᶍᶁᶋᶈᵎᵆḎḥ ḤḤ ḞḝᵊᵅᶉḈᶎᶇḅḚḖḜḟḞ ḝ Ḙ Ḝ Ḛ ḗḦḀḋḁḛḗ ḟ ḛ ḐḕḒḇᵇᶀᵂᵒᵃᵗ ᶍ ᶉ Ḩ ḕ Ḗ ḡ ḉ ᶈ Ḏ ḑ Ḓ Ḁ Ḣ ḙ ḇ ḣ ḓ Ḯ ḋ Ḕ Ḧ ḭ Ḫ ḅ Ḉ ᶋ ᶁ ḁ ᶛ ᶇ ᶆᵕᵘᵀᶂᶆ ᶎ ᶂ ḩ Ḃ ᶚᵏ ᶀ ᶖᵚᶘᶊᶏ ᵁ ᵈ ᵉ ᵍ ᵋ ᵏᵄᵝᶒᶚ ᶊ ᶜ ᵆ ᵂ ᵒ ᵘ ᵇ ᵀ ᵄ ᵊ ᵃ ᵚ ᵎ ᵖ ᵝ ᵢ ᵥ �ᵢᵌ�י �יᵣ ᵣᵖᶜ� � ᵔ �
openai/whisper-large-v3 (pseudo-label, language set)
null
tr
0.33765
0.06653
0.00027
openai/whisper-large-v3 via faster-whisper
wrong_language
zxx_Latn
1
false
false
false
1
158.9
amh_u007869
1a_lessons
amh
train
FSI
FSI - Amharic Basic Course - Volume 1 - Unit 03.mp3
null
null
PD-US
17 U.S.C. § 105
1.2
20
0.842
251.03
252.23
ስስ ሰሰ ሲሳሲ ሴርሻምራልለማመሚሁህላሜሎሮሡሆሊረሕሃሓሖሉሱሒሗሂሀሬሠሄሟሙሩሔሣሢሪሇሼሴ ር ማ ም መ ሚ ለ ላ ል ሮ ሙ ሜ ህ ሁ ሉ ሆ ሃ ሇ ሎ ሀ ሂ ሖ ሪ ረ ሗ ሊ ሳ ሠ ሡ ሓ ሢ ሄ ሩ ሕ ሑ ሒ ራ ሻ ሣ ሥ ሱ ሞ ሬ ሟ ሏ ሽ ሌ ሔ ሦ ሼ ሯሥሞሯ ሤሏሽሌሷሑሤ ሾሹሾ ሹ ሷ ሐሺሿሦሐ ሺ ሸሸ ሿ ሧ ሶሧሶ �, �ه �
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.46436
0.06091
0.00208
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.6495
false
true
false
1
165.83
amh_u003794
1a_lessons
amh
train
FSI
34b.mp3
null
null
PD-US
17 U.S.C. § 105
1.61
10.6
1
97.08
98.69
ስስ ሰሰ ሲሳሲ ሳ ርር መሻሚልለማምመ ም ማ ሚ ለ ረረ ል ሮሼሮ ህሴራሁህ ላሜሩሎላ
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.24707
0.16589
0.00163
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.9064
false
true
false
1
32.3
amh_u002683
1a_lessons
amh
train
FSI
26c.mp3
null
null
PD-US
17 U.S.C. § 105
4.61
19.7
0.766
46.04
50.65
ለለ መልመ ል ላላ ምህም ሚርስሚ ማማ ር ረሰረ ስ ሰ ሲሳሩራሁ ህ ራ ሳ ሮሲ ሁሮ ሣሪሬሜሽሆሎሊሕሠሡሣ ሊ ሠ ሥሉሯሓሻሢሄሗሖሀሱሃሟሒሥ ሆ ሗ ሉ ሖ ሪ ሕ ሓ ሜ ሀ ሩ ሃ ሄ ሌሂሙሼሑሤሏሇሔሐሌ ሐ ሤ ሑ ሇ ሂ ሬ ሒ ሎ ሢ ሙ ሟ ሡ ሏ ሞሹሞ ሽ ሔ ሦ ሱ ሴ ሻ ሯ ሼ ሷ ሹ ሸ ሺ ሧ ሿ ሶ ሾ �
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.89795
0.01697
0.00429
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.7622
false
true
false
1
39.91
amh_u000320
1a_lessons
amh
train
FSI
13c.mp3
null
null
PD-US
17 U.S.C. § 105
1.9
13.5
1
99.47
101.37
ስስ ሰሰ ሲሳሲ ሳ ርር መሻራሚልለማምመ ም ማ ለ ሚ ረሮሬህላሁሜሎሕሆሊሠሉሡረ ል ሠ ሮ ሡ ላ ሣሸ ህ ሜ ሁ ሉ ሀ ሙሽሪሀሓሃሖሗሄሟሢሒሱሂሩሣ ሕ ሪ ራ ሖ ሩ ሗ ሊ ሥ ሓ ሎ ሃ ሑ ሇ ሂ ሆ ሢ ሒ ሞ ሦ ሬ ሟ ሄ ሻ ሤ ሌ ሏ ሔ ሱ ሐሹ ሽ ሯሥሞሏሦሑሇሔሌሷሤሙ ሼሯ ሹሺሼ ሷ ሴሴ ሺ ሸሾሿሐ ሾ ሿ ሧ ሶሧሶ �, �ه�,
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.85645
0.00131
0.00103
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.5983
false
true
false
1
104.74
amh_u001325
1a_lessons
amh
train
FSI
18a.mp3
null
null
PD-US
17 U.S.C. § 105
2.04
14.4
1
311.42
313.46
ስስ ሰሰ ርሳር መሲመ ምሚልም ማለማ ሚ ለ ሲ ረረ ል ላሮህሁራሎሜሬሠሢሪሕሊሩሡላ
openai/whisper-large-v3 (pseudo-label, language set)
null
yo
0.2854
0.06372
0.06726
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.7176
false
true
false
1
24.51
amh_u003263
1a_lessons
amh
train
FSI
30b.mp3
null
null
PD-US
17 U.S.C. § 105
1.86
18.9
1
517.99
519.85
እእ ካካ ነነ አንናኈአ ና ክኛኡኚ኎ክ ኩኜኝኖኞኩ ኡ ከኘኙኗኍኢኲኣ ኰከ ኤኰ ኢ ኝ ኖ ኞ ን ኘ ኦኵኤ ኚ ኛ ኜ ኮኔኦ ኣኬ ኪኋኪ ኙ ኗ ኈ ኑ ኔ ኬኀ ኵ ኲ ኋ ኳ ኟ ኀኮ ኎ ኁ ኊ ኍ ኉ ኧ ኼ ኹ ኻ ኂኻኳኂ ኽ ኅኁኹኟ኉ኼኧኅ ኏ኆ኏ ኆ ኒኒ ኇኊ኱ኑኃ ኄኇ ኿ ኯ኿ኽኯ ኾኄ ኃኾ ኱ ኌኸ ኷኷ ኸኺ ኺኌ ኶኶ ኴ �, �
openai/whisper-large-v3 (pseudo-label, language set)
null
mt
0.1792
0.00084
0.1311
openai/whisper-large-v3 via faster-whisper
unverified
tir_Ethi
0.4013
false
true
false
1
104.84
amh_u008213
1a_lessons
amh
train
FSI
FSI - Amharic Basic Course - Volume 1 - Unit 05.mp3
null
null
PD-US
17 U.S.C. § 105
1.06
25.2
0.83
1,188.93
1,189.99
ስስ ሰሰ ሲሲ ሳሳ ሚርሚ ር ምማም መመ ማ ለረለ ልሩል ረ ሮሮ ሡራሜሎሁህሆላሕሠሡ ህ ሠ ሜ ሩ ሖሪሖ ሁ ሪ ራ ሢ ሙሬሗሀሓሃሢሄሊሣሟሒሱሽሉሷሥሂሌሇሻሞሑሙ ሆ ሣ ላ ሀ ሞ ሃ ሗ ሕ ሓ ሑ ሎ ሉ ሂ ሟ ሒ ሥ ሊ ሬ ሇ ሻ ሦ ሱ ሄ ሽ ሌ ሔሸሏሔ ሼሤሦሐሼ ሏ ሺሯሾሹሺ ሷ ሤ ሐ ሸ ሯ ሧ ሹ ሴ ሾ ሿሴሿ ሶሧሶ �, �
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.24182
0.12067
0.00006
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.7083
false
true
false
1
184.91
amh_u006172
1a_lessons
amh
train
FSI
49a.mp3
null
null
PD-US
17 U.S.C. § 105
5.33
19.9
0.897
223.82
229.15
ለለ ልል መመ ሚህሚ ማማ ምርም ር ረሰረ ስስ ሰ ራላሜራ ህ ላ ሲሠሲ ሁሳሮሁ ሆሎሬሩሆ ሊሖሊ ሠ ሜ ሖ ሮ ሉሪሉ ሓሻሕሓ ሕ ሪ ሳ ሀሣሥሀ ሌሗሃሡሄሢሒሱሂሽሔሟሌ ሇሦሯሤሇ ሗ ሥ ሣ ሬ ሃ ሐሹሼሑሏሐ ሄ ሑ ሎ ሦ ሒ ሂ ሩ ሢ ሙሙ ሡ ሟ ሤ ሞሞ ሱ ሏ ሔ ሴ ሻ ሯ ሼ ሽ ሷሿሴሷ ሺሾሺ ሸሸ ሹ ሾ ሧሧ ሿ ሶሶ �, �ه�,
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.24634
0.03412
0.01813
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.6419
false
true
false
1
37.34
amh_u007966
1a_lessons
amh
train
FSI
FSI - Amharic Basic Course - Volume 1 - Unit 04.mp3
null
null
PD-US
17 U.S.C. § 105
1.06
13.6
0.915
242.95
244.01
ስስ ሰሰ ሴርሲሩሳምልለሎማሚህመላሆሊሜሗሁራሕሃረሄሉሠሖሓሒሱሪሀሮሡሣሟሬሙሑሦሢሌሴ ር ሲ ም ማ መ ሚ ለ ሠ ሙ ሜ ሡ ሮ ሢ ል ላ ሪ ሆ ህ ሖ ሃ ሁ ሳ ሩ ሎ ሉ ሟ ሞ ረ ሗ ሀ ሻ ሕ ሂሻሞሥሂ ሄ ሑ ሇ ሓ ሒ ራ ሊ ሥ ሣ ሬ ሏ ሌ ሱ ሦ ሼ ሽ ሔሷሏሇሽሔ ሷ ሤሯሤ ሯ ሐሹሿሼሧሐ ሹ ሺሾሺ ሸሸ ሧ ሾ ሿ ሶሶ �, �
openai/whisper-large-v3 (pseudo-label, language set)
null
tr
0.17908
0.03311
0.00004
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.5894
false
true
false
1
184.91
amh_u000011
1a_lessons
amh
train
FSI
12a.mp3
null
null
PD-US
17 U.S.C. § 105
3.61
17.9
0.861
161.27
164.88
ለለ መልላሚሁህመ ል ላ ማሊርምማ ሚ ም ር ሰሰ ስስ ረሲረ ሲ ራሎሳራ ሁ ህ ሥሉሜሬሮሆሀሕሖሃሗሒሱሓሪሄሥ
openai/whisper-large-v3 (pseudo-label, language set)
null
si
0.60742
0.0062
0.07843
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.5765
false
true
false
1
18.01
amh_u000724
1a_lessons
amh
train
FSI
15c.mp3
null
null
PD-US
17 U.S.C. § 105
1.89
16.9
0.915
3.88
5.77
ለለ መመ ሚስሚ ማልምማ ል ም ርሰር ስ ሰ ረህሲረ ሲ ሳሮሳ ላላ ህ ሮ ሖሽሜራሁሪሎሬሩሢሠሡሕሖ ሁ ሠ ሜ ሙሻሟሓሥሣሱሆሊሗሀሄሃሉሼሙ ሆ ሡ ሉ ሀ ሊ ሪ ሃ ሇ ሕ ሓ ሎ ሂ ሗ ሢ ሑ ሩ ሣ ሥ ራ ሄ ሟ ሞሹሯሂሒሺሞ ሱ ሤሤ ሒ ሬ ሌ ሐሌሔሐ ሦሴሦ ሽ ሔ ሏ ሴ ሯ ሻ ሸሑሸ ሼ ሷሏሇሿሾሷ ሺ ሾ ሹ ሿ ሧሧ ሶሶ �, �
openai/whisper-large-v3 (pseudo-label, language set)
null
si
0.1593
0.07526
0.00608
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.6537
false
true
false
1
103.7
amh_u006326
1a_lessons
amh
train
FSI
50b.mp3
null
null
PD-US
17 U.S.C. § 105
2.77
23
0.736
330.02
332.79
ስስ ሰሰ ርሳር ሳ ሲሲ ምማሚልለም መረመ ማ ሚ ረ ለ ሠሮሯሁራህ ል ሣሬሜሎላሣ ላ ሮ ሡሪሊሠ ሁ ሜ ሖሖ ሪ ራ ሬ ሉሩሉ ህሻሢሀሗሆሂሕሃሡ ሆ ሥሓሒሱሄሏሇሥ ሗ ሕ ሓ ሩ ሢ ሙሹሙ ሑሼሟሑ ሇ ሟ ሃ ሀ ሊ ሦ ሞሽሤሞ ሂ ሤ ሐሦሌሔሷሐ ሎ ሄ ሒ ሏ ሌ ሔ ሱ ሻ ሯ ሼ ሷ ሽ ሴ ሹ ሸ ሺ ሧ ሾ ሿሺሿ ሶሾሴሧሸሶ �
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.8667
0.01246
0.00612
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.8501
false
true
false
1
69.68
amh_u009431
1a_lessons
amh
train
FSI
FSI - Amharic Basic Course - Volume 1 - Unit 11.mp3
null
null
PD-US
17 U.S.C. § 105
2.54
22.3
0.843
1,069.61
1,072.15
ስስ ሰሰ ርሳር መምመ ም ማሚልለማ ሚ ረሲረ ለ ሲ ል ሮሮ ሣራሜሽሡሁህላሎሬሠሣ ላ ሳ ሠ ሖሩሖ ሁ ራ ሜ ሪሪ ህ ሩ ሕሊሻሕ ሆሱሆ ሑሼሢሓሃሗሀሉሟሄሒሥሔሂሌሇሑ ሓ ሥ ሗ ሃ ሊ ሀ ሉ ሬ ሞሞ ሂ ሟ ሙሹሙ ሎ ሡ ሢ ሒ ሦሯሤሏሐሦ ሄ ሤ ሇ ሱ ሐ ሏ ሻ ሔ ሌ ሯ ሼ ሽ ሧ ሺሺ ሷሿሴሾሷ ሾ ሸሸ ሴ ሹ ሿ ሶሧሶ �, �
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.60986
0.00357
0.07397
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.7422
false
true
false
1
77.17
amh_u001181
1a_lessons
amh
train
FSI
17b.mp3
null
null
PD-US
17 U.S.C. § 105
2.87
18.9
0.927
368.68
371.55
ስስ ሰሰ ርሳር መማሚልለምመ ለ ም ማ ሚ ረሲረ ል ሲ ላሮህላ
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.28198
0.02426
0.01052
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.9814
false
true
false
1
13.24
amh_u008870
1a_lessons
amh
train
FSI
FSI - Amharic Basic Course - Volume 1 - Unit 08.mp3
null
null
PD-US
17 U.S.C. § 105
1.06
16.8
1
1,569.73
1,570.79
ḍḍ ḏḘḏ Ḙ ḠḠ ḚḉḥḜḝḚ
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.13708
0.02873
0.00001
openai/whisper-large-v3 via faster-whisper
wrong_language
zxx_Latn
1
false
false
false
1
16.98
amh_u008224
1a_lessons
amh
train
FSI
FSI - Amharic Basic Course - Volume 1 - Unit 05.mp3
null
null
PD-US
17 U.S.C. § 105
3.13
26.2
0.754
1,256.11
1,259.24
ስስ ሰሰ ሳሳ ሲሲ ሴርመራ ር መ ምሮሚልም ሚ ማማ ለረለ ል ረ ሮ ሠሻሜሩሁህሽሆሎሱሠ ህ ሣሬሖሕሊሪ ሁ ላሣ ሖ ሪላ ሩ ሜ ራሉሓሃሗ ሆ ሗሒሂሀሄሢሏሔሐሇሥሡሟሑሙሾሯሦሞሴ ሓ ሦ ሬ ሀ ሥ ሉ ሑ ሙ ሃ ሕ ሢ ሎ ሞ ሟ ሡ ሂ ሤ ሒ ሇ ሊ ሻ ሱ ሄ ሼ ሐ ሌ ሔ ሽ ሏ ሯ ሺ ሷሤሿሼሺሸሷ ሾ ሸ ሹሹ ሿ ሧ ሶሧሌሶ �, � �
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.77832
0.02368
0.00032
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.7835
false
true
false
1
63.26
amh_u004091
1a_lessons
amh
train
FSI
36c.mp3
null
null
PD-US
17 U.S.C. § 105
2.51
12.8
0.928
32.08
34.59
ለለ ልል መመ ላሚህላ ምማም ማ ሚ ረርረ ር ስሰሲስ ሰ ሲ ራሮሳራ ሁሜሩሪሁ ሜ ሖሖ ሮ ሠሠ ህ ሣሬሻሕሆሎሱሊሗሉሓሃሡሄሀሟሒሽሥሣ ሊ ሥ ሪ ሉ ሬ ሀ ሆ ሗ ሃ ሓ ሇ ሐሂሙሇሏሢሌሑሯሦሼሤሞሴሾሔሐ ሄ ሌ ሕ ሦ ሑ ሎ ሂ ሩ ሒ ሙ ሢ ሡ ሳ ሟ ሞ ሤ ሏ ሻ ሔ ሱ ሯ ሼ ሴ ሽ ሧ ሹ ሷ ሺ ሸ ሿ ሾ ሶሹሿሺሸሷሧሶ �, �
openai/whisper-large-v3 (pseudo-label, language set)
null
si
0.36548
0.01241
0.00358
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.7088
false
true
false
1
78.09
amh_u005960
1a_lessons
amh
train
FSI
47c.mp3
null
null
PD-US
17 U.S.C. § 105
3.51
26.6
0.65
192.58
196.09
ለለ መልላሚህመ ል ላ ማሊማ ሚ ረምርረ ም ር ሰሰ ስስ ሲሲ ሳሳ ሮሁራሎሜሬሮ ሁ ሜ ሖሉሻሆሀሖ ህ ሠሠ ሣሩሓሢሪሕሥሱሗሣ ሌ ሉ ሩ ሙሒሯሂሄሟሡሞሙ ሆ ሗ ሊ ራ ሕ ሪ ሓ ሡ ሢ ሃሹሿሤሽሃ ሄ ሇሔሴሺሑሼሸሷሏሌሶሇ ሂ ሥ ሐሐ ሀ ሑ ሒ ሦሾሦ ሎ ሤ ሟ ሞ ሬ ሔ ሱ ሏ ሻ ሯ ሼ ሽ ሷ ሺ ሸ ሴ ሹ ሿ ሾ ሶ ሧ �, �ه �
openai/whisper-large-v3 (pseudo-label, language set)
null
tr
0.22644
0.16833
0.0004
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.8218
false
true
false
1
56.41
amh_u008402
1a_lessons
amh
train
FSI
FSI - Amharic Basic Course - Volume 1 - Unit 06.mp3
null
null
PD-US
17 U.S.C. § 105
2.58
13.4
1
1,374.76
1,377.34
ለለ መልመ ል ላላ ሚህሚ ምማም ማ ስርስ ር ሰሰ ሲሲ ረሳሮራሁ ሁረ ህ ራ ሖሎሜሩሬሪሣሡሢሠሽሗሖ ሆሊሻሕሱሆ ሊ ሮ ሗ ሉሉ ሠ ሜ ሃሃ ሀሹሯሓሥሀ ሄሒሾሂሄ ሇሌሟሼሤሞሑሔሇ ሓ ሐሐ ሂ ሕ ሌ ሑ ሪ ሎ ሣ ሳ ሩ ሢ ሙሙ ሱ ሒ ሡ ሥ ሬ ሟ ሞ ሻ ሤ ሏ ሦ ሼ ሽ ሔ ሯ ሺሏሦሿሴሺ ሷሷ ሾ ሹ ሸሸ ሴ ሿ ሧሧ ሶሶ �, �
openai/whisper-large-v3 (pseudo-label, language set)
null
ps
0.22803
0.00805
0.03311
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.6724
false
true
false
1
75.97
amh_u001541
1a_lessons
amh
train
FSI
19b.mp3
null
null
PD-US
17 U.S.C. § 105
1.31
13.6
1
249.15
250.46
ለለ መስመ ልሚል ሚ ማሰማ ምም ረርረ ር ስ ሰ ሲሲ ላሳሮራሁህላ
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.71826
0.01049
0.00528
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.7456
false
true
false
1
30.53
amh_u000194
1a_lessons
amh
train
FSI
12c.mp3
null
null
PD-US
17 U.S.C. § 105
1.76
11.4
1
27.41
29.17
ልለል ለ መመ ሚስሚ ምርም ማማ ር ስ ረሰሲሡሳራሬሜሩሠሢረ ሰ ሠ ሲ ሡ ሮሮ ሳ ሜ ላላ ህህ ሆሻሪሖሎሱሁሆ ሁ ሖ ሪ ራ ሩ ሣ ሉሉ ሃሣሗሥሓሦሕሼሟሞሃ ሗ ሊሊ ሓ ሕ ሀሙሽሀ ሇሂሒሿሾሄሇ ሄ ሂ ሥ ሐሌሔሏሐ ሎ ሑ ሒ ሙ ሢ ሞ ሦ ሟ ሬ ሌ ሔ ሤ ሏ ሻ ሱ ሼ ሯሤሹሯ ሽ ሷሑሺሴሧሷ ሺ ሸሸ ሴ ሧ ሹ ሾ ሿ ሶሶ �, �
openai/whisper-large-v3 (pseudo-label, language set)
null
es
0.45361
0.00396
0.00174
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.5528
false
true
false
1
111.36
amh_u005245
1a_lessons
amh
train
FSI
42e.mp3
null
null
PD-US
17 U.S.C. § 105
3.18
18.5
0.868
35.39
38.57
ለለ መመ ሚሚ ልምርስሲል ማማ ም ር ስ ሰረሰ ሡህራሳሮሁሜሬሩሎሡ ህ ረ ሮ ሲ ሠላሹሱሆሊሠ ሁ ሜ ሖሖ ላ ሆ ሙሪሼሢሽሓሣሕሿሟሃሗሀሉሄሻሥሂሒሯሔሌሇሑሙ ሀ ሪ ሉ ሊ ሳ ሃ ሗ ሎ ሂ ሣ ሩ ሢ ሑ ሇ ሞ ራ ሕ ሓ ሒ ሥ ሟ ሄ ሻ ሦ ሱ ሏ ሌ ሬ ሔ ሤ ሐሞሷሏሤሦሺሴሾሸሐ ሽ ሯ ሼ ሧ ሹ ሷ ሾ ሸ ሺ ሴ ሿ ሶ �, �ه �
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.46362
0.04813
0.00003
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.5462
false
true
false
1
61.95
amh_u007402
1a_lessons
amh
train
FSI
58c.mp3
null
null
PD-US
17 U.S.C. § 105
1.85
17.6
0.881
173.33
175.18
እእ ነነ አንኚአ ኡካናኛኡ ና ኩክኜኝኞ ኢኩ ን ከኖ኎ኙኢ ኖ ክ ኰኘኔኑ ኙ ኤከ ኚ ኝ ኞኰ ኛ ኦ ኘ ኜ ኣኈኗ ኗኲኅ ኪኵኣ ካ ኔ ኑኊ ኮ ኈ ኵ ኬኤ ኲ ኹ ኳ ኯኬ ኟ ኋኀ ኧ ኎ ኁ ኀኋ ኒ ኂ ኊኍ ኍኒኁኳኂኮኦኆኄ኏኉ ኅኧኃ ኆ ኄ ኏ ኿ኹኇኪ ኼኽ ኽ�
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.87305
0.01236
0.00001
openai/whisper-large-v3 via faster-whisper
unverified
tir_Ethi
0.2718
false
true
false
1
83.78
amh_u000024
1a_lessons
amh
train
FSI
12a.mp3
null
null
PD-US
17 U.S.C. § 105
4.56
18.5
0.882
266.27
270.83
እናውለልት ነደስችተታቶቲቸባቹቱቷቭቻቢበ አንያርሰዚቼቴብቫቡቩቺቮቪቤቚዳምቀቖቍክድመቨቅቈቕቬቁቛቆ቎ቂቦቝማሁላቿቜቘህሚሊቾቯቓቌቔቐቊራነ ን ና ኩረሀሉቧቋሎሆዎወዝይኛዛግገእ ኡሮሲሜጋጊዋዘጎአ ይ ካየጥሳጫጛኝጉሬሖሪጁዊጣሠ቏዁ጀዉዬጠሃ቞ቄቒ቟ቑቃዲጚጅዖጕሓጇጓሕጆጄጏ
openai/whisper-large-v3 (pseudo-label, language set)
null
am
0.33398
0.0048
0.33398
openai/whisper-large-v3 via faster-whisper
wrong_language
cmn_Hani
0.0458
false
false
false
1
34.21
amh_u008106
1a_lessons
amh
train
FSI
FSI - Amharic Basic Course - Volume 1 - Unit 05.mp3
null
null
PD-US
17 U.S.C. § 105
1.86
22.1
0.86
261.44
263.3
ያዚደያ ይየይ የ ድድ ደ ዳዳ ዲዲ ዹዱውዝዎዛወዥዬዪ ዩዮዯስሚልለማመምህርሠሁሉራሬሜሕዩ ዴዹ ዸዼሮሰረሀላሆሊሗዕሎሓዽሳሩሡዖሖዣሲሟሪ ጥንናሽዸ ዪሣዜዻሶሢሃሱሄሌሇሥሞሯትችታብቫቲቚተቮቪባቸሼቶቷቱቭቻቹቩቺቼእ�ቬበቀቛቁቴቤቡቢቅቯዋቝቖቍገቘቦ቞ቨቾቧቂቜቐቕቓቃቆቄቊቈጥ ው ዛ �
openai/whisper-large-v3 (pseudo-label, language set)
null
mt
0.88916
0.00009
0.05774
openai/whisper-large-v3 via faster-whisper
wrong_language
zxx_Latn
0.2558
false
false
false
1
88.17
amh_u000074
1a_lessons
amh
train
FSI
12b.mp3
null
null
PD-US
17 U.S.C. § 105
1.31
15.2
1
147.24
148.55
ስስ ሰሰ ሳሳ ሲሲ ርመለልሚህር መ ምማም ማ ሚ ለ ረረ ል ላላ ሠሽሮሁራሎሜሩሆሊሱሠ ህ ሮ ሡሬሖሕሃሡ ሁ ሜ ሖ ሪሓሒሪ ሆ ሣሻሢሂሗሀሄሉሟሣ ሓ ሩ ሢ ሙሹሥሇሼሙ ሗ ሃ ሕ ሊ ራ ሉ ሀ ሥ ሑ ሇ ሂ ሄ ሎ ሒ ሤ ሞ ሟ ሬ ሌ ሐሞሯሏሌሔሷሑሤሦሐ
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.84424
0.01038
0.00043
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.7851
false
true
false
1
114.5
amh_u002042
1a_lessons
amh
train
FSI
23b.mp3
null
null
PD-US
17 U.S.C. § 105
2.41
17.8
0.88
21.08
23.49
ልለል ለ መመ ሚህሚ ማማ ምርረም ር ረ ሰሰ ስስ ሲሲ ሳሳ ሡላራሜሽሮ ህ ሠሻሬሠ ላ ሮሖሎሁሩሕሆሊሗ ሣሣ ሁ ሀሪሉሓሃሡ ሜ ሖ ሪ ራ ሉ ሬ ሗሀ ሊ ሟሙሥሢሂሱሒሟ ሢ ሙ ሥ ሓ ሕ ሃ ሆ ሑ ሩ ሎ ሤ ሞሼሯሤሄሏሇሔሷሑሞ ሄ ሦሴሹሦ ሂ ሌሸሶሾሿሌ ሇ ሐሧሐ ሒ ሏ ሔ ሱ ሧ ሻ ሼ ሯ ሽ ሷ ሴ ሹ ሸ ሺ ሿ ሶ ሾ �, �ه �
openai/whisper-large-v3 (pseudo-label, language set)
null
de
0.31665
0.01034
0.00056
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.6906
false
true
false
1
82.16
amh_u008347
1a_lessons
amh
train
FSI
FSI - Amharic Basic Course - Volume 1 - Unit 06.mp3
null
null
PD-US
17 U.S.C. § 105
1.11
14.4
0.928
940.68
941.79
Яхин вэ тау к'Аннэн?
openai/whisper-large-v3 (pseudo-label, language set)
null
de
0.18579
0.03177
0.00007
openai/whisper-large-v3 via faster-whisper
wrong_language
evn_Cyrl
0.9032
false
false
false
1
18.02
amh_u009437
1a_lessons
amh
train
FSI
FSI - Amharic Basic Course - Volume 1 - Unit 11.mp3
null
null
PD-US
17 U.S.C. § 105
2.37
19.3
0.958
1,115.33
1,117.7
ስስ ሰሰ ርሳር ለማሚልለ መምመ ማ ም ሚ ረሲረ ል ሲ ራህላሁራ ላ ሠሮሯሎሜሩሊሕሆሱሉሬሠ ሜ ሮ ሖሖ ሁ ህ ሥሓሻሀሗሂሄሃሢሽሔሣሪሒሾሥ ሕ ሪ ሳ ሓ ሬ ሣ ሀ ሉ ሊ ሗ ሃ ሆ ሂ ሇ ሩ ሎ ሙሹሼሡሇሞሟሏሌሷሑሤሙ ሡ ሌ ሐሸሺሴሿሶሐ ሢ ሑ ሒ ሤ ሄ ሦ ሱ ሔ ሟ ሞ ሯ ሻ ሏ ሽ ሷ ሼ ሹ ሸ ሺ ሴ ሿ ሾ ሧ ሶ �, �ه �
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.32861
0.01662
0.09418
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.7849
false
true
false
1
83.12
amh_u007535
1a_lessons
amh
train
FSI
59c.mp3
null
null
PD-US
17 U.S.C. § 105
1.65
15
1
401.35
403
ለለ መመ ልል ሆሚሆ ሚ ምህም ማማ ርረር ስሰስ ሰ ረ ሲሲ ሜራሜ ላላ ሠሖሎሁሳሩሢሬሠ ሁ ሖ ሮሗሮ ህ ራ ሦሪሼሡሣሟሥሱሃሽሓሕሒሊሹሻሉሸሄሀሂሇሿሾሑሌሔሐሤሞሯሦ ሕ ሪ ሗ ሉ ሥ ሬ ሓ ሐ ሤ ሞ ሊ ሳ ሡ ሎ ሀ ሃ ሟ ሙሙ ሂ ሩ ሒ ሢ ሑ ሣ ሇ ሻ ሄ ሏ ሌ ሔ ሱ ሯ ሼ ሽ ሷሏሺሷ ሺ ሸ ሧ ሹ ሴ ሿ ሾ ሶሧሴሶ �, �
openai/whisper-large-v3 (pseudo-label, language set)
null
te
0.45923
0.02673
0.00018
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.7786
false
true
false
1
118.79
amh_u004509
1a_lessons
amh
train
FSI
38b.mp3
null
null
PD-US
17 U.S.C. § 105
2.79
22.7
0.778
315.11
317.9
ስስ ሰሰ ርሳር መሚልለማምመ ማ ረሲረ ሚ ም ለ ሲ ል ራህላሜራ ላ ሳ ሖሎሮሁሆሊሕሠሬሪሄሉሱሖ ሠ ሮ ሜ ሬ ሪ ሁ ሣ ሀ ሉ ህ ሊ ሡሻሀሓሩሂሗሢሡ ሢ ሃሃ ሆ ሗ ሕ ሎ ሓ ሇ ሩ ሂ ሙሒሥሣሟሞሯሽሔሌሷሏሹሑሼሙ ሄ ሑ ሐሸሤሴሺሦሾሇሐ ሤ ሒ ሥ ሱ ሌ ሦ ሞ ሟ ሔ ሏ ሻ ሯ ሼ ሷ ሴ ሽ ሺ ሸ ሹ ሿሿ ሾ ሧ ሶሧሶ �, �
openai/whisper-large-v3 (pseudo-label, language set)
null
pa
0.375
0.02512
0.0008
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.742
false
true
false
1
70.25
amh_u005563
1a_lessons
amh
train
FSI
45b.mp3
null
null
PD-US
17 U.S.C. § 105
1.9
24.6
0.789
77.52
79.42
ḍḍ ḏḘḉḠḏ ḋḝḜḚḋ Ḙ ḝ Ḡ Ḛ ḥḥ Ḝ ḤḤ ḗḎḖḈḀḁḛ
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.83203
0.0473
0
openai/whisper-large-v3 via faster-whisper
wrong_language
zxx_Latn
1
false
false
false
1
20
amh_u003479
1a_lessons
amh
train
FSI
32b.mp3
null
null
PD-US
17 U.S.C. § 105
1.01
17
1
70.24
71.25
ḵḨᵍḍᵁḥḠḵ ḥ Ḡ ḤḤ ḡḘḉḝḜḡ ḢḚḎḀḋḛḅḏḇḁᵉᵋᵈḈᵆḃḞḖḐḆᵇᵊᵎᵒḟ�ᶍᶁᶋማᵃᵅᵘᵗḒᵖᵀᵂᵛᶈᶉᶎᶖᶀᶊᶚᵝᵚᶠᵏᵄᵌᵐᵕᵦᵜᵠᶇᶜᶘᶝᶛᴇᴛọመᴍግ�ንḕᶫᵔḌḂḗᵢᵓḦḢ Ḩ Ḛ ḝ Ḙ Ḝ Ḟ ḟ ḗ ḍ ḛ Ḑ ḏ ḉ Ḁ Ḗ Ḏ ḕ Ḓ Ḧ ḣ ḋ Ḯ ḑ ḇ ḁ ḭ Ḇ Ḃ ḃ Ḉ ḓ ḅ Ḍ Ḕ Ḋ ḄḊḄ ḫḣḫ ḪḔᵣᵮḮᵥ�ᵽᵨᶏ�
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.41431
0.03976
0.00029
openai/whisper-large-v3 via faster-whisper
wrong_language
zxx_Latn
1
false
false
false
1
191.09
amh_u002177
1a_lessons
amh
train
FSI
24b.mp3
null
null
PD-US
17 U.S.C. § 105
1.27
15
1
206.57
207.84
ስስ ሰሰ ሲሳሲ ሳ ሴርረሚልለምመማራሎህላሜሩሁሆሊሠሕሃሡሉሱሖሓሒሗሂሬሪሀሮሟሄሥሢሌሣሯሦሞሙሑሤሇሴ ር ም ማ መ ሚ ለ ሠ ረ ሮ ሡ ሪ ሜ ሙ ሖ ሢ ል ላ ሟ ህ ሁ ሉ ሃ ሩ ሎ ሞ ራ ሆ ሥ ሗ ሕ ሀ ሣ ሓ ሻ ሬ ሇ ሂ ሒ ሦ ሑ ሼ ሱ ሄ ሊ ሽ ሏ ሌ ሷ ሤ ሯ ሐሻሏሽሔሹሼሷሐ
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.33496
0.05341
0.00125
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.652
false
true
false
1
133.07
amh_u008505
1a_lessons
amh
train
FSI
FSI - Amharic Basic Course - Volume 1 - Unit 07.mp3
null
null
PD-US
17 U.S.C. § 105
1.14
26.4
0.623
393.7
394.84
ስስ ሰሰ ሲሲ ሳሳ ሴርሚልለምመማሡህላሁሽሓሎሮራሕሊረሀሆሉሱሃሜሻሬሠሖሗሂሼሙሒሪሄሟሢሣሩሔሌሇሦሞሯሥሴ ር ም ማ ሚ መ ለ ላ ሮ ሙ ሁ ሜ ሉ ል ህ ሆ ሃ ሀ ሎ ሡ ሪ ረ ሖ ሠ ሗ ሕ ሥ ሟ ሞ ራ ሻ ሢ ሂ ሩ ሣ ሬ ሓ ሦ ሼ ሒ ሽ ሑ ሇ ሱ ሊ ሌ ሏ ሄ ሷ ሔ ሤ ሯ ሐሹ ሹሷሏሑሺሤሐ
openai/whisper-large-v3 (pseudo-label, language set)
null
lv
0.23401
0.00648
0.00006
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.7399
false
true
false
1
152.63
amh_u001286
1a_lessons
amh
train
FSI
18a.mp3
null
null
PD-US
17 U.S.C. § 105
1.93
12.3
0.959
16.31
18.24
ስስ ሰሰ ሳሳ ሲሲ ርር ምመሚልም መ ማለማ ሚ ረረ ለ ሠሻሜራሁህሆላሎሓሬሠ ል ራ ሮሮ ላ ሡሪሖሀሕሊሣሩሢሄሉሡ ሁ ሪ ሖ ሩ ሜ ሣ ሙሯሹሗሂሃሱሒሽሟሥሞሑሦሤሼሙ ህ ሥ ሬ ሓ ሕ ሗ ሉ ሊ ሐሸሴሇሿሾሌሔሐ ሆ ሃ ሀ ሑ ሇ ሂ ሒ ሎ ሄ ሦ ሢ ሞ ሟ ሤ ሌ ሏ ሻ ሔ ሱ ሯ ሼ ሴ ሽ ሷሏሷ ሺሺ ሸ ሹ ሾ ሧ ሿ ሶሧሶ �, �
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.54639
0.00386
0.00536
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.7164
false
true
false
1
101.55
amh_u004132
1a_lessons
amh
train
FSI
36c.mp3
null
null
PD-US
17 U.S.C. § 105
1.93
15.5
0.87
336.34
338.27
እእ ነነ አንና ናኛኚ኎ ኢክኲኝ ኡካኈኘኡ ኩኖኍኆአ ን ከኦኵኩ ኦ ኰኰ ክ ኤኬኜኂ ኝከ ኞኹ ኖ ኣኙኞ ኚ ኘ ኛ ኙ ኜ ኮ ኗኀኗ ኑ ኪኁ ካ ኔኤ ኧ ኟ ኵ ኬ ኈ ኳ ኲ ኀ ኹኔ ኎ኮኣ ኒ ኂኻ ኋኢ ኁ኉ ኯኒኪ ኅ ኿ኊ ኍ ኊኋ ኆ ኼ ኻኅኑኽ ኽኃ ኏኏ ኾ ኱ኼኧኟኇ ኷ኾ኱ ኶኿ ኺ ኸ኷ ኃኳኯ ኉ኸ ኄኺኄ ኇ኶ ኴ ኌ �, � mრጋ��,
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.85693
0.02733
0.00005
openai/whisper-large-v3 via faster-whisper
unverified
tir_Ethi
0.4425
false
true
false
1
104.15
amh_u003100
1a_lessons
amh
train
FSI
29b.mp3
null
null
PD-US
17 U.S.C. § 105
1.33
17.8
1
412.53
413.86
እእ ነነ የንናዳስሳትችታኡቲቶተቸቷቱቻቹቼቴቢበቡሰደለማቩቺባቤብቮቪቭቫቚቬቖቀቦቍቈቆቁቾቯቿቛልርምቂቅቧቨቜቝመ቎ቕ቞ቘሚሁላቋህረሄቄሊቓቃ቟቙቗ቊሜሎሉቑቔቐቒ቉� ን ና ይክዛውይ አኑዚያሮሲሀሆዎወዋ�ኢሃሖሬራዮዝ዁ዬዓሓ ኡ ደ ድኩኖዲድ ያ ዯአ ዚ ዮ ዲ ዝ ዛ ዳ ዬ ው ዠየ ወ ዥዻዥ ዻ ዜዱዉ
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.91357
0.00681
0.00021
openai/whisper-large-v3 via faster-whisper
unverified
tir_Ethi
0.114
false
true
false
1
130.08
amh_u005935
1a_lessons
amh
train
FSI
47b.mp3
null
null
PD-US
17 U.S.C. § 105
1.15
11.1
1
391.63
392.78
ለስሰሚርሲመሳምማልለ ል መ ስ ም ሚ ማ ሲ ረሮሯሩራሁሜህሆላሎረ ር ሰ ሮ ሡሪሬሠሢሡ ሁ ሠ ሜ ሙሹሻሟሓሽሖሱሼሗሕሿሃሣሙ ላ ሪ ሖ ሩ ህ ሣ ሢ ሉ ሃ ሆ ሎ ሀ ሳ ሗ ሞ ሟ ሥ ሂ ሕ ሑ ሇ ራ ሓ ሦ ሒ ሻ ሬ ሱ ሄ ሽ ሊ ሷ ሯ ሤ ሏ ሼ ሔ ሸ ሌ ሺ ሐሂሉሥሀሞሦሒሺሊሷሏሄሇሌሔ�, ሧ ሹ ሴ ሿ ሾ ሶሤሧሴሑሸሐ �,ሾሶ �
openai/whisper-large-v3 (pseudo-label, language set)
null
tr
0.4209
0.01385
0.00008
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.641
false
true
false
1
172.17
amh_u004087
1a_lessons
amh
train
FSI
36c.mp3
null
null
PD-US
17 U.S.C. § 105
1.99
23.2
0.754
4.95
6.94
ለለ መሰመ ልርል ሚሲሚ ሰ ስስ ር ምሮም ማማ
openai/whisper-large-v3 (pseudo-label, language set)
null
ur
0.81006
0.01875
0.00002
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.9946
false
true
false
1
14.07
amh_u005766
1a_lessons
amh
train
FSI
46b.mp3
null
null
PD-US
17 U.S.C. § 105
1.39
17.8
0.892
225.44
226.83
ስስ ሰሰ ሳሳ ሲሲ ሴርረምልሚሁሽመሎለማህላሡራሕሆሊሠሜሬሩሖሮሟሪሀሓሢሱሒሼሗሃሣሙሾሯሦሞሴ ር ም ማ መ ሚ ለ ሠ ሙ ሜ ሮ ሡ ረ ሪ ል ሢ ህ ላ ሩ ሖ ሁ ሟ ሉ ሃ ሎ ሞ ሆ ሻ ሗ ሕ ሂሻሥሂ ራ ሄ ሇ ሥ ሓ ሬ ሀ ሣ ሑ ሒ ሦ ሱ ሏ ሊ ሷ ሽ ሯ ሼ ሌ ሔሉሹሏሄሇሌሔ ሺ ሤ ሐሺሤሑሿሧሐ ሾ ሹ ሸሷሸ ሧ ሿ ሶሶ �, �
openai/whisper-large-v3 (pseudo-label, language set)
null
sv
0.33032
0.03271
0.00003
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.5939
false
true
false
1
141.01
amh_u008850
1a_lessons
amh
train
FSI
FSI - Amharic Basic Course - Volume 1 - Unit 08.mp3
null
null
PD-US
17 U.S.C. § 105
1.11
13
1
1,431.54
1,432.65
ስስ ሰሰ ሳሳ ሲሲ ርር ማለምሚልማ
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.61865
0.00017
0.03128
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.4375
false
true
false
1
18.92
amh_u001813
1a_lessons
amh
train
FSI
21b.mp3
null
null
PD-US
17 U.S.C. § 105
2.29
18
0.943
225.24
227.53
ስስ ሰሰ ሲሳሲ ሳ ርር ሚልለሚ መማምመ ማ ም ለ ረረ ል ላሮህሁራሜሎሬሠሢሪሀላ
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.2981
0.05603
0.16211
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.751
false
true
false
1
21.4
amh_u008951
1a_lessons
amh
train
FSI
FSI - Amharic Basic Course - Volume 1 - Unit 09.mp3
null
null
PD-US
17 U.S.C. § 105
1.17
17.8
0.923
352.82
353.99
ስስ ሰሰ ርሳር ሴሲሚልምለማመህላሆሁሎረሄሊራሕሠሜሬሩሡሢሮሓሪሖሞሱሉሼሴ መ ም ማ ለ ሚ ሲ ረ ላ ል ሮ ሳ ሜ ሙሻሽሥሀሗሣሟሙ ሁ ሪ ሖ ሠ ሃሹሯሃ ህ ሆ ሊ ሉ ሀ ሇ ሓ ሎ ሂ ሕ ሗ ሑ ሒ ራ ሢ ሄ ሩ ሡ ሥ ሣ ሟ ሞ ሻ ሬ ሌ ሏ ሱ ሐሂሒሏሇሌሔሑሤሦሧሐ
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.7417
0.0003
0.02661
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.5909
false
true
false
1
134.19
amh_u003762
1a_lessons
amh
train
FSI
34a.mp3
null
null
PD-US
17 U.S.C. § 105
1.07
18.9
0.925
367.75
368.82
ስስ ሰሰ ሲሳሲ ሴርረምልለመማሚህላራሁሎሄሊሜሆሉሱሠሕሃሡሀሮሖሓሻሢሽሪሬሣሩሟሼሗሂሌሴ ር ማ ም መ ለ ሚ ረ ሮ ሠ ሳ ሡ ል ላ ህ ሪ ሁ ሜ ሙሒሦሞሥሯሤሿሑሙ ሃ ሣ ሖ ሩ ሢ ሉ ሗ ሎ ሀ ሟ ሞ ራ ሆ ሂ ሄ ሊ ሕ ሒ ሇ ሻ ሑ ሓ ሏ ሥ ሬ ሌ ሱ ሦ ሽ ሼ ሯ ሤ ሔሷሏሇሔ ሷ ሺሾሹሐሺ ሾ ሐ ሹ ሧ ሿ ሸ ሶሧሸሶ �, �ه �
openai/whisper-large-v3 (pseudo-label, language set)
null
he
0.38794
0.0752
0.00111
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.6827
false
true
false
1
185.98
amh_u001411
1a_lessons
amh
train
FSI
18b.mp3
null
null
PD-US
17 U.S.C. § 105
1.07
13.1
1
422.26
423.33
ስስ ሰሰ ሲሲ ሴርሳሩምሡሚልለመማህሽረሪላሆሎሱሁሜራሕሊሮሠሄሀሓሬሟሗሖሉሴ ር ም ማ መ ሚ ሙሹሻሢሂሙ ል ለ ሮ ሜ ሡ ላ ሠ ሪ ሢ ሖ ሩ ሳ ሟ ሁ ህ ሉ ሃ ሆ ሎ ሗ ሻ ረ ሕ ሂ ሱ ሇ ራ ሞ ሑ ሓ ሥ ሬ ሒ ሽ ሣ ሀ ሷ ሄ ሏ ሼ ሯ ሦ ሹ ሊ ሸ ሌ ሔ ሧ ሺ ሤ ሿ ሾ ሐሃሣሥሞሏሒሑሇሼሦሔሌሷሤሐ ሶሯሾሧሿሺሸሶ �, �
openai/whisper-large-v3 (pseudo-label, language set)
null
es
0.28442
0.09161
0
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.7212
false
true
false
1
183.18
amh_u002187
1a_lessons
amh
train
FSI
24b.mp3
null
null
PD-US
17 U.S.C. § 105
1.92
15.6
1
265.57
267.49
ስስ ሰሰ ርሳር ሳ ሲሲ ምማሚልም መረመ ማ ሚ ረ ለሮለ ል ሣሩህላሁራሎሜሬሠሡሣ ላ ሮ ሡ ሠ ሜ ሖሪሖ ሁ ሪ ራ ህ ሩ ሢ ሙሻሢሀሆሊሗሃሕሉሱሓሟሄሥሽሔሂሒሾሇሌሷሑሼሯሦሞሹሿሸሤሴሙ ሆ ሥ ሓ ሕ ሗ ሊ ሐሺሶሏሐ ሑ ሇ ሀ ሉ ሃ ሒ ሎ ሂ ሟ ሞ ሬ ሄ ሻ ሦ ሱ ሤ ሌ ሏ ሔ ሷ ሯ ሼ ሽ ሹ ሺ ሸ ሴ ሧ ሾ ሿ ሶ �, �
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.854
0.00146
0.00374
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.8096
false
true
false
1
101.56
amh_u001220
1a_lessons
amh
train
FSI
17b.mp3
null
null
PD-US
17 U.S.C. § 105
2.31
25
0.823
649.47
651.78
ለለ መስሰሚልርመ ል ሚ ማሲማ ር ምረም ስ ረ ሰ ሲ ሮ ሡሳሜራህሁሩሎሪላሟሬሠሮሡ ላ ሜ ሙሻሖሱሆሀሗሓሕሢሣሥሞሯሦሒሽሉሙ ህ ሠ ሖ ሗ ሁ ሕ ሓ ሇሸሴሤሄሂሃሑሼሔሊሷሏሌሇ ሆ ሐሺሧሿሾሹሐ ሌ ሀ ሉ ሊ ሃ ሒ ሑ ሪ ሎ ሂ ራ ሄ ሣ ሳ ሟ ሢ ሞ ሩ ሥ ሱ ሬ ሦ ሻ ሔ ሤ ሏ ሽ ሯ ሼ ሷ ሴ ሺ ሸ ሹ ሿ ሾ ሧ ሶሶ �, �
openai/whisper-large-v3 (pseudo-label, language set)
null
da
0.43921
0.22791
0.00094
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.7669
false
true
false
1
84.85
amh_u000476
1a_lessons
amh
train
FSI
14b.mp3
null
null
PD-US
17 U.S.C. § 105
2.77
20.4
0.845
301.45
304.22
ለለ መልመ ል ሚስሚ ማሰምማ ም ርረር ስ ረ ሰ ሲህሲ ላላ ሳሮሳ ህ ሜሻራሜ ሁሁ ሖሪሬሎሆሊሕሠሽሓሃሡሢሀሉሟሗሖ ሠ ሮ ሣሩሣ ራ ሕ ሪ ሗ ሉ ሊ ሬ ሓ ሙሹሯሥሂሱሒሼሑሄሇሔሐሌሙ ሆ ሥ ሩ ሡ ሀ ሃ ሦ ሢ ሎ ሑ ሒ ሂ ሇ ሄ ሻ ሟ ሞሞ ሱ ሤ ሌ ሐ ሏ ሔ ሽ ሯ ሼ ሴ ሹ ሷ ሺ ሸሤሏሦሺሾሿሴሷሧሸ ሾ ሧ ሶሶ ሿ �, �ه �
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.89355
0.02202
0.00216
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.5123
false
true
false
1
71.84
amh_u008123
1a_lessons
amh
train
FSI
FSI - Amharic Basic Course - Volume 1 - Unit 05.mp3
null
null
PD-US
17 U.S.C. § 105
1.67
14.6
0.946
476.3
477.97
ስስ ሰሰ ርሳር ሳ ሲሲ መሚልለምመ ም ማማ ለ ሚ ረረ ል ላላ ሣህሮሁራሎሜሆሊሕሠሬሄሉሓሗሖሀሱሡሃሢሒሩሂሪሇሣ ህ ሠ ሮ ሖ ሗ ሁ ራ ሓ ሜ ሪ ሉ ሩ ሊ ሬ ሙሻሙ ሆ ሃ ሕ ሀ ሥ ሂ ሇ ሎ ሑ ሒ ሢ ሄ ሟ ሡ ሌ ሐሹሥሔሏሽሟሌሤሑሞሯሦሐ ሦ ሤ ሱ ሏ ሞ ሯ ሔ ሧ ሻ ሼ ሴ ሽ ሷሼሧሿሾሺሷ ሾ ሺ ሸሸ ሹ ሿ ሶሴሶ �, �ه �
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.67969
0.00031
0.03833
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.7272
false
true
false
1
119.16
amh_u001593
1a_lessons
amh
train
FSI
19c.mp3
null
null
PD-US
17 U.S.C. § 105
3.54
20.8
0.836
194.03
197.57
ስስ ሰሰ ርሳር መሲመ ለማለ ማ ምሚልም ል ረረ ሚ ሲ ሳ ላላ ህህ ሆሁራሜሎሆ ሁ ሠሮሠ ሜ ራ ሖሽሪሕሊሬሓሉሱሖ ሗሒሩሢሄሀሗ ሕ ሥሻሣሥ ሓ ሣ ሮ ሉ ሊ ሪ ሬ ሀ ሃሹሼሂሃ ሄ ሇሺሯሦሞሟሇ ሟ ሑሸሡሌሔሐሏሑ ሡ ሐ ሩ ሒ ሂ ሦ ሢ ሎ ሞ ሤ ሙሿሤሙ ሱ ሌ ሏ ሔ ሻ ሯ ሽ ሼ ሴ ሷሷ ሹ ሺ ሸ ሧሾሴሧ ሾ ሿ ሶሶ �, �
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.77344
0.00505
0.0082
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.7511
false
true
false
1
55.37
amh_u004906
1a_lessons
amh
train
FSI
41b.mp3
null
null
PD-US
17 U.S.C. § 105
1.68
10.6
0.935
334.91
336.59
ስስ ሰሰ ሲሲ ሳሳ ርር ሴመሻሚልለምማሁህላሎራሕሊሜሉሮሡሄሆሀሓረሬሠሖሗሃሟሪሂሱሒሽሩሔሣሼሢሞሯሦሙሾሥሴ ማ ም መ ለ ሚ ረ ሮ ሠ ሜ ሪ ሬ ሣ ላ ል ሖ ሁ ሉ ህ ሙ ሃ ሎ ሀ ሡ ሞ ራ ሢ ሗ ሊ ሟ ሥ ሩ ሕ ሂ ሻ ሓ ሦ ሒ ሆ ሄ ሱ ሑ ሇ ሼ ሽ ሯ ሤ ሏ ሌ ሔ ሷ ሐሹ ሹሷሏሇሌሤሑሺሐ
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.63525
0.02791
0.00058
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.7749
false
true
false
1
104.17
amh_u002514
1a_lessons
amh
train
FSI
26a.mp3
null
null
PD-US
17 U.S.C. § 105
1.63
14.6
1
260.43
262.06
ለለ መስመ ልል ሚሰሚ ምማም ማ ስ ርረር ረ ሰ ላህላ
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.82178
0.00844
0.00204
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.9892
false
true
false
1
20.25
amh_u007037
1a_lessons
amh
train
FSI
55c.mp3
null
null
PD-US
17 U.S.C. § 105
1.91
15.7
0.942
401.66
403.57
ስስ ሰሰ ሳሳ ሲሲ ለርልለ ር መረመ ል-ሚህምማሁራላሜሎሠሕሊሬሄሆሉሩሀሓሽሗሃሡሖሮሥሣሪሞሱሟሒሯሢሂሻሼሙሑሌሴሤሇሦሐሹሾሸሔሺሷሏሶሧሿ�, ማ ም ረ ሚ ላ ሠ ሮ ህ ሁ ሪ ራ ሜ ሥ ሣ ሬ ሖ ሉ ሀ ሓ ሊ ሟ ሃ ሆ ሎ ሗ ሕ ሄ ሇ ሩ ሂ ሻ ሙ ሑ ሡ ሢ ሞ ሽ ሒ ሱ ሤ ሦ ሼ ሌ ሷ ሏ ሐ ሴ ሯ ሔ ሺ ሧ ሹ ሾ ሸ ሿ ሶ �,
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.40894
0.00247
0.17322
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.5171
false
true
false
1
103.14
amh_u009286
1a_lessons
amh
train
FSI
FSI - Amharic Basic Course - Volume 1 - Unit 10.mp3
null
null
PD-US
17 U.S.C. § 105
1.99
19.1
0.93
1,570.1
1,572.09
ስስ ሰሰ ርሳር ረሚልምለማመራህላሁሩሎረ መ ማ ም ሚ ለ ሲሲ ል ላ ሳ ሮሮ ሁ ሖሖ ሜሻሜ ህ ሠሬሪሆሊሕሠ ሗሉሱሓሒሢሄሗ ሡሣሡ ሕ ራ ሪ ሩ ሣ ሀሹሯሀ ሆ ሬ ሉ ሊ ሥ ሓ ሃሃ ሎ ሂሽሥሂ ሄ ሇ ሤሞሟሇሔሌሏሐሑሤ ሑ ሒ ሐ ሦሼሙሿሦ ሢ ሌ ሟ ሙ ሞ ሱ ሏ ሻ ሔ ሯ ሼ ሹ ሷ ሽ ሺ ሸሺሷሾሸ ሴሴ ሾ ሧሧ ሿ ሶሶ �, �ه �
openai/whisper-large-v3 (pseudo-label, language set)
null
pa
0.31396
0.00196
0.00223
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.8246
false
true
false
1
100
amh_u006743
1a_lessons
amh
train
FSI
53b.mp3
null
null
PD-US
17 U.S.C. § 105
1.35
13.9
1
310.6
311.95
ለለ መስሰመ ልርሳሚል ላማሲምህላ
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.88867
0.0024
0.00878
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.9502
false
true
false
1
14.81
amh_u005235
1a_lessons
amh
train
FSI
42d.mp3
null
null
PD-US
17 U.S.C. § 105
1.15
17.7
0.93
484.99
486.14
ስስ ሰሰ ሲሳሲ ሳ ርር ሚሚ ምም መሻራልለማሁመ ማ ለ ረረ ል ላሮሯሜህላ
openai/whisper-large-v3 (pseudo-label, language set)
null
te
0.24207
0.13367
0.00188
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.9128
false
true
false
1
39.13
amh_u005254
1a_lessons
amh
train
FSI
42e.mp3
null
null
PD-US
17 U.S.C. § 105
4.18
18.7
0.825
89.43
93.61
እናውን ክክ ከኩካልለእ አኛያነአ ነ ካ ኰኘስኚኈሰሚሳመኖማምህርረላሁራችኔሊሎ኎ሉሆኆዎዚትሓሃሮሲሗቲቫወተታቻቸሬቶቷቱቹሜቭብሖቚደይዳኝቼቮባበሠቀሕዲዛቪቤቢቖዋቩቺሀኞሪዝ዁ቴሩሄቡቯቾቅቦቆቁድቿሼሣሌሇሂቕቂሱሡዊቛኡሢዕሻሟቈዓቍቨቧ቞ቜሒቘቬሑቄሏቝኑዱዉቓኦሞዖኜዐ቟የ
openai/whisper-large-v3 (pseudo-label, language set)
null
ps
0.24365
0.06354
0.02509
openai/whisper-large-v3 via faster-whisper
wrong_language
zxx_Latn
0.0708
false
false
false
1
36.84
amh_u001947
1a_lessons
amh
train
FSI
22b.mp3
null
null
PD-US
17 U.S.C. § 105
2.58
17.9
1
316.49
319.07
ስስ ሰሰ ርሳር መሲመ ለማሚልለ ል ማ ምረም ሚ ረ ሲ ሳ ላሮህላ
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.69434
0.00316
0.02692
openai/whisper-large-v3 via faster-whisper
repetitive
tig_Ethi
0.9656
false
true
false
5
15.5
amh_u005492
1a_lessons
amh
train
FSI
44c.mp3
null
null
PD-US
17 U.S.C. § 105
2.15
17.1
0.94
6.85
9
እናዳለእ ነነ አንደልስመአ ና ኩኈርሰኖሳችትቶታኡቲተቸቷቻቼቱቹቺቴቭቮብባቢ኎ቫበቚቪቤቡሲምቩቿቖቍቦቆኛቅቨቾቯቁቛሚውክኝማካሁላቈቀቬቘቂቜቝኩ ን ክ ከኘሜህሮረሀሎኚሪሬሣራኢቕቧ቞቟ቓቋሊሉሖኔ቎ሃቔኑቄቐቊ቗ቃሆ቏ቑሓሕሄሌቒ቙቉ቌሱሂሒሇቇሠሏሢሩሟሡገግጋጥሗጛጕኜጇጊጀጓ
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.42798
0.10168
0.05072
openai/whisper-large-v3 via faster-whisper
wrong_language
cmn_Hani
0.1405
false
false
false
1
72.09
amh_u003369
1a_lessons
amh
train
FSI
31b.mp3
null
null
PD-US
17 U.S.C. § 105
2.47
23.4
0.854
257.32
259.79
ለለ መልመ ል ሚስሚ ስ ምርሰም ር ማማ ሰ ረላህራሲረ ላ ሠሳሮሁሩሬሜሠ ህ ሣሉሻሢሽሡሎሕሱሆሊሗሖሄሓሃሣ ሁ ሮ ሲ ሉ ሊ ሳ ሆ ሖ ሜ ሀ ሃ ሙሹሪሀሂሥሼሟሒሯሦሔሌሇሐሏሑሴሤሙ ሌ ሥ ሪ ራ ሓ ሕ ሗ ሐ ሇ ሂ ሩ ሑ ሎ ሡ ሄ ሢ ሒ ሟ ሞሷሞ ሱ ሤ ሦ ሬ ሻ ሔ ሏ ሽ ሯ ሼ ሷ ሴ ሹ ሸ ሺ ሿሿ ሾሺሾ ሶሧሸሶ ሧ �
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.80273
0.06287
0.00234
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.6193
false
true
false
1
78.14
amh_u004821
1a_lessons
amh
train
FSI
40c.mp3
null
null
PD-US
17 U.S.C. § 105
1.88
20.5
0.771
85.65
87.53
ስስ ሰሰ ሲሳሲ ሳ ርር ምመሚልም መ ማለማ ሚ ለ ል ረሮሁህላራሢሎሬረ ላ ሮ ሡሪሖሜሻሡ ሁ ሠሽሠ ህ ሣሩሕሊሱሆሉሓሼሗሀሄሃሟሒሂሌሔሣ ሀ ሪ ራ ሩ ሖ ሗ ሜ ሉ ሃ ሊ ሤ ሂ ሆ ሙሙ ሎ ሓ ሇ ሢ ሑ ሕ ሥ ሬ ሄ ሻ ሟ ሌ ሱ ሒ ሞሥሏሞ ሽ ሦ ሯሤሹሇሑሦሾሯ ሼ ሔ ሐሸ ሴሺሿሐ ሹ ሏ ሸሴ ሺ ሷሷ ሾ ሿ ሧሧ ሶሶ �, �
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.62842
0.04733
0.00002
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.6912
false
true
false
1
104.26
amh_u007228
1a_lessons
amh
train
FSI
57b.mp3
null
null
PD-US
17 U.S.C. § 105
2.06
18.2
0.942
5.39
7.45
መስሰሚርሳመ ስ ር ምም ሚ ማሲማ
openai/whisper-large-v3 (pseudo-label, language set)
null
ur
0.28833
0.06287
0.00003
openai/whisper-large-v3 via faster-whisper
unverified
tir_Ethi
0.6525
false
true
false
1
9.71
amh_u009382
1a_lessons
amh
train
FSI
FSI - Amharic Basic Course - Volume 1 - Unit 11.mp3
null
null
PD-US
17 U.S.C. § 105
2.2
17.3
0.945
704.97
707.17
ስስ ሰሰ ሲሳሲ ሳ ርር ሚልለምሚ መማመ ም ማ ለ ረረ ል ሮሮ ሣሩህራሁሜሽሡሎሆላሕሠሢሪሀሓሣ ህ ላ ሠ ሜ ሖሻሬሖ ሁ ሡ ሉሸሯሉ ሆ ሃሱሊሗሃ ሄሒሿሂሄ ሇሴሙሇ ሂ ሗ ሊ ሀ ሓ ሪ ራ ሕ ሑ ሙ ሩ ሎ ሢ ሞሹሥሔሌሏሑሟሼሤሞ ሒ ሱ ሐሺሾሐ ሌ ሥ ሬ ሤ ሟ ሦ ሻ ሼ ሔ ሯ ሏ ሺ ሽ ሷሦሷ ሹ ሸ ሴ ሾ ሧሧ ሿ ሶሶ �, �
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.29004
0.00947
0.0002
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.7783
false
true
false
1
89.09
amh_u000427
1a_lessons
amh
train
FSI
14a.mp3
null
null
PD-US
17 U.S.C. § 105
2.47
21.3
0.834
209.41
211.88
ስስ ሰሰ ርሳር ሳ ሲመምሚልለማራሎረህላሁሩሊሮሠሬሜሪሄሉሱሲ ማ ም መ ለ ሚ ረ ሮ ል ላ ራ ህ ሁ ሜ ሖሽሆሖ ሪ ሠ ሬ ሣሻሢሓሹሣ ሓ ሩ ሥሀሯሕሂሗሃሡሒሟሥ ሗ ሃ ሕ ሉ ሊ ሀ ሆ ሂ ሇ ሄ ሙሙ ሎ ሑ ሒ ሢ ሡ ሞሼሴሞ ሱ ሤሺሤ ሟ ሌሷሔሌ ሔ ሐሸሑሾሏሇሐ ሏ ሦሦ ሴ ሻ ሯ ሼ ሽ ሧሿሧ ሹ ሺ ሸ ሷ ሾ ሿ ሶ �, �ه � �
openai/whisper-large-v3 (pseudo-label, language set)
null
tr
0.52197
0.00477
0.05334
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.7484
false
true
false
1
80.97
amh_u008193
1a_lessons
amh
train
FSI
FSI - Amharic Basic Course - Volume 1 - Unit 05.mp3
null
null
PD-US
17 U.S.C. § 105
1.96
19.4
0.929
1,077.4
1,079.36
ስስ ሰሰ ርሳር መሲመ ምሚም ሚ ማማ ልለሜራል ለ ረረ ሲ ራ ሮሩህሁሮ ሁ ሳ ሖሎሬሪሆላሽሟሠሢሣሡሓሕሖ ህ ሠ ሜ ሩ ላ ሕ ሪ ሆ ሗሊሻሱሀሹሯሂሗ ሓ ሣ ሥሙሒሃሉሼሥ ሢ ሃ ሊ ሬ ሉ ሀ ሙ ሂ ሑ ሇ ሞሞ ሎ ሟ ሡ ሒ ሦሔሄሏሇሷሑሦ ሄ ሌሺሤሴሾሌ ሱ ሤ ሐሿሧሐ ሽ ሔ ሧ ሻ ሯ ሏ ሴ ሼ ሷ ሺ ሸሸ ሹ ሿ ሾ �, ሶሶ �
openai/whisper-large-v3 (pseudo-label, language set)
null
cs
0.57861
0.01076
0.00017
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.6802
false
true
false
1
100
amh_u002161
1a_lessons
amh
train
FSI
24b.mp3
null
null
PD-US
17 U.S.C. § 105
1.64
21.2
0.848
91.07
92.71
ለለ መሰሚስርሳመ ልምማል ላላ ም ሚ ማ ስ ሰ ር ሲሲ ረህራረ ሁሩሮሯሁ ህ ሜሻሜ ሆሖሎሪሬሡሢሠሣሕሱሼሗሽሓሆ ሊሊ ሮ ሖ ሠ ሉሉ ሀሹሙሸሟሥሀ ሇሃሒሷሞሴሄሂሇ ሂ ሃ ሕ ሗ ሓ ሐሌሔሐ ሄ ሑ ራ ሪ ሎ ሩ ሙ ሒ ሳ ሡ ሣ ሥ ሞ ሟ ሢ ሬ ሦ ሻ ሱ ሌ ሤ ሏ ሔ ሽ ሯ ሼ ሷ ሹ ሺ ሸ ሴ ሿሺሿ ሾሏሤሦሾ ሶሑሧሶ ሧ �
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.41968
0.0183
0.00037
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.6711
false
true
false
1
117.68
amh_u002287
1a_lessons
amh
train
FSI
24c.mp3
null
null
PD-US
17 U.S.C. § 105
3.04
19.9
0.875
458.24
461.28
ስስ ሰሰ ርሳር ለሲመልለ መ ምማም ማ ሚሚ ል ረረ ላራህላ
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.35449
0.00388
0.1897
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.9855
false
true
false
2
11.84
amh_u007435
1a_lessons
amh
train
FSI
59b.mp3
null
null
PD-US
17 U.S.C. § 105
2.66
21.2
0.85
57.78
60.44
ለለ መላልሚሁህመ ል-ማምርረስራሰሡሳሩሲሠሜሬሮሯሊሕሣሪሎሱሆሀሓሖሄሗሉሟሢሞሥሴሤሼሙሃሒሽሔሐሂሻሺሑሦሸሹሌሇሶሾሷሿሏሧ�, ማ ም ር ስ ላ ሚ ህ ሁ ሉ ሊ ሆ ሇ ሜ ሃ ሀ ሄ ሌ ሓ ረ ሖ ሰ ሲ ሮ ሎ ሂ ሪ ራ ሕ ሗ ሳ ሠ ሥ ሣ ሩ ሡ ሢ ሙ ሟ ሞ ሑ ሒ ሻ ሬ ሦ ሽ ሏ ሐ ሤ ሱ ሔ ሴ ሯ ሼ ሷ ሹ ሺ ሧ ሾ ሿ ሸ ሶ �,
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.38208
0.0147
0.12408
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.5922
false
true
false
1
74.06
amh_u002821
1a_lessons
amh
train
FSI
28a.mp3
null
null
PD-US
17 U.S.C. § 105
1.67
16.6
1
329.8
331.47
ስስ ሰሰ ርሳር ረምመሚልለማራህረ መ ም ሚ ሲሲ ማ ለ ል ራ ላሮሯሁሜሬሎሪሆሊሟላ
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.5166
0.00515
0.00954
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.8115
false
true
false
1
29.94
amh_u004736
1a_lessons
amh
train
FSI
40a.mp3
null
null
PD-US
17 U.S.C. § 105
8.15
25.5
0.647
323.3
331.45
ለለ ልል ላላ መመ ሚህሚ ምማም ማ ርረር ስሰስ ሰ ረ ሲሳሲ ህ ራሉራ ሁሁ ሖሮሬሜሎሕሆሊሠሖ ሊ ሠ ሜ ሉ ሮ ሆ ሗሪሥሓሣሢሀሗ ሌሃሡሒሱሄሟሼሩሔሂሑሦሯሽሌ ሇሻሹሤሞሾሐሏሇ ሐ ሥ ሕ ሓ ሪ ሀ ሬ ሃ ሣ ሦ ሳ ሎ ሂ ሄ ሏ ሟ ሙሙ ሑ ሡ ሒ ሞ ሢ ሤ ሔ ሱ ሩ ሼ ሯ ሻ ሽ ሴ ሸ ሹ ሷ ሺ ሶሺሸሿሴሷሧሶ ሾ ሧ ሿ �
openai/whisper-large-v3 (pseudo-label, language set)
null
pa
0.35669
0.08807
0.00339
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.6419
false
true
false
1
23.68
amh_u007798
1a_lessons
amh
train
FSI
FSI - Amharic Basic Course - Volume 1 - Unit 02.mp3
null
null
PD-US
17 U.S.C. § 105
1.1
13.5
1
229.22
230.32
ስስ ሰሰ ሳሳ ሲሲ ርር መማሚልለምመ ሴረህላሩሎሡሁራሕሊሮሠሜሬሟሄሀሓሻሢሱሆሉሴ ም ማ ለ ረ ሚ ላ ሮ ል ህ ሁ ሜ ሃሪሂሗሃ ሆ ሠ ሪ ራ ሓ ሩ ሖሙሖ ሗ ሕ ሥ ሣሣ ሀ ሙ ሊ ሡ ሉ ሎ ሬ ሟ ሞ ሂ ሄ ሻ ሢ ሇ ሱ ሑ ሒሒ ሽ ሌ ሐሽሥሞሏሹሔሌሷሇሼሑሯሦሐ
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.59229
0.00942
0.01003
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.8387
false
true
false
1
141.82
amh_u001938
1a_lessons
amh
train
FSI
22b.mp3
null
null
PD-US
17 U.S.C. § 105
1.7
21.6
0.847
255.22
256.92
ḍḍ ḏḠḏ Ḡ ḡḘḜḡ ḥḝḚḥ Ḙ ḝ Ḛ ḗḉḎḀḋḈḅḇḁḗ Ḝ ḛḖḕḊḌḂḄḆḃḛ
openai/whisper-large-v3 (pseudo-label, language set)
null
ru
0.2981
0.05734
0.00002
openai/whisper-large-v3 via faster-whisper
wrong_language
zxx_Latn
1
false
false
false
1
28.24
amh_u005508
1a_lessons
amh
train
FSI
44c.mp3
null
null
PD-US
17 U.S.C. § 105
2.39
23.3
0.782
109.81
112.2
ስስ ሰሰ ርሳር ሴሲመምሚልለማራህላሁሩሎረሬሠሜሮሡሀሱሆሊሣሪሉሓሕሖሄሗሃሢሌሻሟሙሼሯሦሞሥሽሴ ሲ ም ማ ለ መ ሚ ረ ሮ ሳ ላ ሪ ሜ ል ህ ሁ ሠ ሖ ሙ ሬ ሉ ሩ ሡ ሃ ሀ ሣ ሥ ራ ሢ ሞ ሻ ሟ ሎ ሗ ሂሹሂ ሆ ሊ ሦ ሒ ሓ ሕ ሇ ሱ ሄ ሑ ሽ ሏ ሌ ሼ ሯ ሤ ሔሷሒሏሇሔ ሷ ሺሤሑሿሾሺ ሹ ሐሸሧሐ ሾ ሸ ሧ ሿ ሶሶ �, �ه �
openai/whisper-large-v3 (pseudo-label, language set)
null
tr
0.26343
0.03766
0.00042
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.7775
false
true
false
1
83.26
amh_u000226
1a_lessons
amh
train
FSI
12c.mp3
null
null
PD-US
17 U.S.C. § 105
1.98
19.7
0.899
262.39
264.37
እእ አንናስልነለማመምህርላሚሳሰረሁራሊሜሎሄሉሓአ ነ ና ን ካኈኛኡሲሌሮሣሠሆሀሱሡትችተታቲቶቷቸባቹቱቻቺቢ኎ኚብቫቼቚኞቩቮቭቪቤበቡኖቍቖሬቀቴቦቨቯቾቂቛቅቕ቞ሪሖውሕሃቁቧቬሩሂሗቈ� ኩኔሻሢቆቘቜቝቔቑ቟ቓቃሼሇሐቋቿ቗ሒቊ቎ቄቐክኝካ ኡ ከኩ ክ ኯኜኘኗዚያወ቏ቒሽሏሟሔዛዋ�ግይ
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.52148
0.0014
0.07397
openai/whisper-large-v3 via faster-whisper
wrong_language
cmn_Hani
0.0781
false
false
false
1
80.3
amh_u006595
1a_lessons
amh
train
FSI
52b.mp3
null
null
PD-US
17 U.S.C. § 105
5.83
21.7
0.751
290.15
295.98
ስስ ሰሰ ርሳር ለሲመልለ መ ማሚህምማ ም ረረ ሚ ል ሲ ላላ ህ ሠሁሜራሎሕሆሊሠ ሕ ራ ሁ ሥሮሓሉሬሄሀሗሏሂሖሽሪሣሡሢሌሟሩሒሱሃሯሥ ሆ ሮ ሖ ሪ ሉ ሣ ሜ ሊ ሀ ሳ ሓ ሇ ሐሻሼሞሑሇሐ ሗ ሃ ሄ ሌ ሬ ሂ ሎ ሑ ሏ ሙሹሿሔሤሙ ሡ ሢ ሒ ሩ ሦሺሴሦ ሟ ሤ ሞ ሯ ሱ ሻ ሔ ሴ ሽ ሼ ሸ ሺ ሷ ሹ ሿ ሾ ሧ ሶሷሾሧሶ �, �ه �
openai/whisper-large-v3 (pseudo-label, language set)
null
am
0.30029
0.07361
0.30029
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.7183
false
true
false
1
33.96
amh_u005968
1a_lessons
amh
train
FSI
48a.mp3
null
null
PD-US
17 U.S.C. § 105
3.5
17.8
0.883
81.36
84.86
ለለ መልርመ ል ሚሰሚ ማስምማ ም ር ስ ሰ ረሲህራሜሳሎሁሓረ ላላ ህ ሲ ሮሊሮ ሁ ሳ ሜ ሖሽሆሀሬሕሪሢሣሠሖ ሊ ሠ ሉሉ ሆ ራ ሕ ሗሻሥሗ ሌሃሩሡሄሼሒሱሂሹሯሾሔሌ ሇሴሟሞሙሿሑሦሤሇ ሓ ሪ ሐሐ ሀ ሥ ሣ ሩ ሃ ሑ ሂ ሄ ሦ ሎ ሙ ሒ ሱ ሤ ሡ ሢ ሞ ሟ ሬ ሔ ሻ ሯ ሏ ሼ ሽ ሷሏሺሷ ሺ ሸሸ ሴ ሹ ሾ ሿ ሧሧ ሶሶ �, �
openai/whisper-large-v3 (pseudo-label, language set)
null
sw
0.43823
0.06311
0.00874
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.6395
false
true
false
1
56
amh_u004884
1a_lessons
amh
train
FSI
41b.mp3
null
null
PD-US
17 U.S.C. § 105
4.46
18.6
0.827
190.82
195.28
እናደለልመላርስም ነነ የሰትችታተያሚባቶቲቸቹቱቷቭቻቼቴቺ አንዚሳህማሁሲረቿራበሬብቫአ ን ኩእ ና ደ ክክ ካድካ ኝኛ ድ ከኩ ኖኝ ይዳኖ ኞሮቮሼሎቚዲይ ያ ወኈውቢቁቾቪቤቀቩቬሉቖቍወ
openai/whisper-large-v3 (pseudo-label, language set)
null
am
0.32007
0.05307
0.32007
openai/whisper-large-v3 via faster-whisper
plausible
amh_Ethi
0.4649
true
true
false
1
24.44
amh_u006184
1a_lessons
amh
train
FSI
49b.mp3
null
null
PD-US
17 U.S.C. § 105
2.02
19
0.96
16.57
18.59
ስስ ሰሰ ሳሳ ሲሲ ርር ምም መሚልለራማህመ ማ ረሮረ ሚ ለ ል ሮ ላላ ሠሻሜሁሪሆሎሕሊሠ ሜ ሡሖሽሟሓሃሡ ሁ ሖ ሪ ራ ህ ሣሩሢሀሗሉሬሣ ሢ ሗ ሃ ሕ ሉ ሩ ሓ ሎ ሀ ሂሒሱሂ ሆ ሊ ሥ ሑ ሇ ሞ ሟ ሙሥሞሯሄሏሌሔ�
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.32886
0.03836
0.00075
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.6605
false
true
false
1
64.36
amh_u008526
1a_lessons
amh
train
FSI
FSI - Amharic Basic Course - Volume 1 - Unit 07.mp3
null
null
PD-US
17 U.S.C. § 105
1.27
15.7
0.929
526.88
528.15
ስስ ሰሰ ሳሳ ሲሲ ርር ሴመሚልለምማህላሁሩሎሆሊረሮሡሄሉሱሠሜራሕሬሓሪሀሗሖሣሟሢሞሙሃሯሦሽሔሂሒሻሴ መ ም ማ ለ ሚ ረ ሮ ሠ ሙ ሪ ሡ ሢ ሜ ሩ ላ ል ህ ሁ ሖ ሆ ሃ ሣ ሀ ሗ ሎ ሉ ሂ ራ ሞ ሓ ሻ ሟ ሄ ሊ ሬ ሥ ሱ ሑ ሒ ሽ ሕ ሇ ሷ ሦ ሼ ሤ ሯ ሌሹሥሏሇሌ ሏ ሺ ሐሤሑሿሼሧሐ ሹ ሔ ሸ ሧ ሾሺሾ ሿ ሶሷሸሶ �, �
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.9502
0.00078
0.00001
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.7065
false
true
false
1
154.33
amh_u000251
1a_lessons
amh
train
FSI
12c.mp3
null
null
PD-US
17 U.S.C. § 105
1.34
14.4
0.94
436.68
438.02
ስስ ሰሰ ሲሳሲ ሳ ርር ምሩም መረሚልመ ማማ ሚ ረ ለሮለ ል ላሻህላ
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.83643
0.03589
0.00004
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.9679
false
true
false
1
31.34
amh_u001754
1a_lessons
amh
train
FSI
21a.mp3
null
null
PD-US
17 U.S.C. § 105
3.17
18.3
0.899
115.56
118.73
ለለ ልል መመ ሚስሚ ምርም ር ስ ሰማሲህራሰ ማ ረላሜሩሳሢረ ሠሠ ሕሪሮሡሁሱሎሬሟሖሣሕ ህ ላ ሲ ሮ ሖ ሁ ራ ሉሉ ሜ ሆሻሥሓሹሯሽሆ ሗሼሗ ሊሊ ሪ ሥ ሓ ሇሒሿሴሃሑሀሄሂሔሌሇ ሑ ሃ ሐሐ ሀ ሣ ሩ ሙሙ ሂ ሒ ሡ ሎ ሄ ሟ ሏ ሳ ሢ ሞሞ ሬ ሦ ሤ ሌ ሔ ሱ ሻ ሼ ሯ ሽ ሷሏሺሤሦሾሷ ሾ ሺ ሸሧሸ ሹ ሴ ሿ ሧ ሶሶ �, �ه �
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.38184
0.04892
0.00349
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.624
false
true
false
1
62.78
amh_u005313
1a_lessons
amh
train
FSI
43b.mp3
null
null
PD-US
17 U.S.C. § 105
3
19
0.85
16.24
19.24
እእ ኩንናካአኩ አ ክክ ካ ነነ ከኯኜኛኝ ና ን ኡኈኚኖ኎ኞኡ ኖ ኙከ ኝኙ ኘኔኘ ኢኋኢ ኚ ኛ ኞ ኤኰኤ ኜ ኰ ኈ ኦኊኮ ኔ ኣኹ ኗኍኣ ኪኁ ኑኲኵኳኦ ኟኂኪ ኊ ኮኗ ኁኻኀኅ ኋ ኵ ኀ ኹኆ኉ኑ ኎ ኅ኱኏ኄኒ ኂ ኆ ኍ ኲ ኉ ኬ ኄ ኇ ኯ ኏ ኒኬኼኧ ኼ ኿ ኳ ኻ ኃኃ ኌኽ ኽኟ ኾ ኺኾ኿ኇኌ ኱ ኶኶ ኷኷ ኸ ኧኺ ኴ �, �
openai/whisper-large-v3 (pseudo-label, language set)
null
sw
0.31055
0.19434
0.02017
openai/whisper-large-v3 via faster-whisper
unverified
tir_Ethi
0.3761
false
true
false
1
64.67
amh_u003785
1a_lessons
amh
train
FSI
34b.mp3
null
null
PD-US
17 U.S.C. § 105
3.09
18
1
45.49
48.58
እእ አንነናኈ኎ኛኘ኏ኍኚኆ ነ ክአ ካካ ኩኜኝኖኁኋኗኅ ና ን ኰክ ኡኡ ኖ ኢኯኀኔኑኙኂኊ኉ኩ ኚ ከኰ ኝ ኤከ ኛ ኘ ኞኮኒኞ ኜ ኦኵኣኤ ኙ ኪኲኢ ኑ ኈ ኮ ኗ ኔ ኣ ኬኪ ኋ ኯ ኀ ኵ ኲ ኹ ኎ ኊ ኳ ኁ ኍ ኅኻኇኌኄኃኦ ኂ ኆኹኽኬ ኏ ኒ ኿኿ ኻ ኟ ኇ ኼ ኄ ኉ ኱኱ ኾ ኽ ኧ ኶ኼኟኳኧኾ�, ኷኷ ኸ ኃ ኌ ኺ ኴ �,
openai/whisper-large-v3 (pseudo-label, language set)
null
ur
0.53125
0.02905
0.00018
openai/whisper-large-v3 via faster-whisper
unverified
tir_Ethi
0.2733
false
true
false
1
62.14
amh_u001624
1a_lessons
amh
train
FSI
20a.mp3
null
null
PD-US
17 U.S.C. § 105
3.21
18.1
0.897
235.06
238.27
ስስ ሰሰ ርሳር መሲመ ለማሚልለ ል ማ ምረም ሚ ረ ሲ ሳ ራህራ ላላ ህ ሠሮሁሜሎሬሠ ሁ ሪሊሩሀሆሓሽሪ ሆ ሮ ሖሉሻሥሱሄሼሕሂሖ ሜ ሥ ሣሸሣ ሊ ሓ ሉ ሬ ሀ ሃሹሃ ሄ ሇሯሢሒሗሇ ሕ ሗ ሎ ሐሺሤሏሌሔሐ ሌ ሑሦሞሟሡሑ ሂ ሩ ሢ ሒ ሦ ሡ ሤ ሙሿሙ ሱ ሟ ሞ ሯ ሔ ሴሴ ሻ ሏ ሽ ሧ ሼ ሷሧሷ ሹ ሺ ሸ ሾ ሿ ሶሾሶ �, �ه �
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.24451
0.00786
0.13501
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.713
false
true
false
1
61.99
amh_u000982
1a_lessons
amh
train
FSI
16b.mp3
null
null
PD-US
17 U.S.C. § 105
2.23
16.4
0.964
108.55
110.78
እእ ካንነናካ ነ አኛኈክአ ና ክ ኩኖ኎ኞኝ ኡኚኅኩ ን ከኡ ኢኋኢ ኖ ኰኘኔኗ ኤኍኆኂኜኙኒኁኳኣ ኛ ኦኊ ኝከ ኞ ኘ ኜ ኚ ኙ ኈ ኗ኉ኤ ኣኰ ኪኀኵኲ ኑ ኵ ኮኦ ኔ ኬኄኹኑኪ ኲኮ ኹ ኁ ኳ ኀ ኋ ኊኬ ኎ ኟ ኍ ኅ ኆ ኂ ኻኼኧ ኼ ኯኻ ኽ ኄ ኏኏ ኿኿ ኾ ኱኱ ኇ ኺ ኒ ኃኽኟኃ ኉ ኸ ኧኇኯ ኶ኾ኶ ኷኷ ኌኸኌ ኴ �, �
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.4895
0.00896
0.00348
openai/whisper-large-v3 via faster-whisper
unverified
tir_Ethi
0.4283
false
true
false
1
87
amh_u005060
1a_lessons
amh
train
FSI
42a.mp3
null
null
PD-US
17 U.S.C. § 105
2.96
18.2
0.919
222.58
225.54
ስስ ሰሰ ሳሳ ሲሲ ርር መማልለምመ ለ ም ማ ሚህሚ ል ረረ ሡላሜራሁሡ ላ ሣሩሎሮሆሊሣ ህ ሠሻሬሪሀሓሕሠ ሁ ሜ ሮ ራ ሪ ሩ ሓ ሕ ሖሉሱሖ ሆ ሗሃሢሄሗ ሊ ሥሙሥ ሉ ሑሒሯሂሟሏሽሔሷሑ ሇሺሤሇ ሢ ሐሹሿሞሴሼሌሶሐ ሎ ሀ ሤ ሃ ሦሸሦ ሄ ሌ ሟ ሙ ሂ ሬ ሒ ሞ ሱ ሏ ሔ ሻ ሯ ሼ ሷ ሽ ሧሾሧ ሹ ሺ ሸ ሴ ሿ ሾ ሶ �, �
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.44336
0.00845
0.04672
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.5438
false
true
false
1
66.22
amh_u004078
1a_lessons
amh
train
FSI
36b.mp3
null
null
PD-US
17 U.S.C. § 105
2.44
18.2
0.893
153.66
156.1
ስስ ሰሰ ሲሳሲ ሳ ርር ምመሚልም መ ማለማ ሚ ረረ ለ ል ሮሮ ሣሩሜራሁህላሎሱሆሊሽሗሃሡሉሠሕሖሀሓሬሪሄሟሣ ሠ ሜ ራ ላ ሩ ሁ ሖ ሗ ሪ ህ ሙሻሢሂሥሞሯሦሒሼሙ ሓ ሥ ሬ ሕ ሉ ሀ ሊ ሃ ሆ ሂ ሇ ሄ ሎ ሑ ሒ ሡ ሢ ሞ ሱ ሦ ሤ ሟ ሔ ሌ ሻ ሯ ሼ ሽ ሏ ሐሹሿሤሏሇሔሌሷሑሾሐ ሷ ሧ ሹ ሴ ሺሧሴሺ ሾ ሸሸ ሿ ሶሶ �, �
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.8916
0.02357
0.00035
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.642
false
true
false
1
80.33
amh_u004641
1a_lessons
amh
train
FSI
39c.mp3
null
null
PD-US
17 U.S.C. § 105
1.82
17.1
0.907
274.19
276.01
ለለ መስመ ልርል ሚሰሚ ማሲምማ
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.78467
0.04257
0.00003
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.9267
false
true
false
2
10.44
amh_u006864
1a_lessons
amh
train
FSI
54b.mp3
null
null
PD-US
17 U.S.C. § 105
1.93
16.8
0.891
172.58
174.51
ስስ ሰሰ ሲሳሲ ሳ ርር ሚረሚ መሻራምማልለመ ም ማ ረ ለ ሠሽሮሯሁህላሜሎሠ ል ሖሴሬሟሪሖ ሜ ራ ሮ ሡሹሩሓሢሡ ሁ ሥሱሆሊሕሥ ህ ሪ ላ ሣሸሣ ሆ ሗሃሉሗ ሉ ሀሷሀ ሃ ሕ ሊ ሐሺሤሞሼሑሂሄሒሾሇሔሌሏሐ ሓ ሬ ሑ ሩ ሙሿሦሙ ሂ ሇ ሢ ሒ ሎ ሞ ሦ ሟ ሄ ሤ ሌ ሔ ሱ ሏ ሻ ሼ ሯ ሽ ሴ ሷ ሺ ሹ ሸ ሿ ሾ ሧ ሶሧሶ �, �
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.67773
0.07141
0.00277
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.621
false
true
false
1
101.55
amh_u008212
1a_lessons
amh
train
FSI
FSI - Amharic Basic Course - Volume 1 - Unit 05.mp3
null
null
PD-US
17 U.S.C. § 105
1.3
13.7
1
1,185.54
1,186.84
ስስ ሰሰ ሲሳሲ ሳ ሚርመሚ መ ምማም ማ ለለ ልረል ር ረ ላሮራሁህላ
openai/whisper-large-v3 (pseudo-label, language set)
null
mt
0.7124
0.00202
0.047
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.9464
false
true
false
1
32.31
amh_u009307
1a_lessons
amh
train
FSI
FSI - Amharic Basic Course - Volume 1 - Unit 11.mp3
null
null
PD-US
17 U.S.C. § 105
1.79
10.4
1
229.39
231.18
ḍḨᵍᵁḍ ḥḠḥ Ḡ ḤḤ ḡḘḉḝḚḎḜḏḁግገḈጋጊጎጉጆጁልለማጌ
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.67578
0.01578
0.00005
openai/whisper-large-v3 via faster-whisper
wrong_language
zxx_Latn
0.9995
false
false
false
1
20.67
amh_u001948
1a_lessons
amh
train
FSI
22b.mp3
null
null
PD-US
17 U.S.C. § 105
2.39
20.7
0.841
323.76
326.15
ስስ ሰሰ ሳሳ ሲሲ ርር ረረ ሴሚልለምመማህላሁራሎሮሡሆሊሜሬሩሄሉሱሠሕሖሀሓሪሂሗሃሢሒሣሟሥሞሯሦሑሤሙሼሴ ም ማ ለ መ ሚ ሮ ሠ ሙ ሡ ል ላ ሁ ህ ሜ ሪ ሖ ሆ ሀ ሉ ሣ ሃ ሊ ራ ሎ ሗ ሢ ሂ ሥ ሓ ሇ ሩ ሕ ሒ ሞ ሟ ሑ ሬ ሄ ሻ ሦ ሱ ሌ ሤ ሏ ሽ ሯ ሼ ሐሻሽሏሇሔሌሹሿሾሐ
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.8125
0.02632
0.00241
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.7137
false
true
false
1
70.29
amh_u005323
1a_lessons
amh
train
FSI
43b.mp3
null
null
PD-US
17 U.S.C. § 105
4.23
16.9
0.903
70.25
74.48
እእ ነነ አንኈናስልለላመህምማሚሁርረሄሊራሉሰሆ኎ሎኖሌሜሠሕሬሮሀሓአ ን ና ኩኊኛደዚትችተታቲቶቷባቸቱብሲሣሳሡቭቻቹቺቮቫቈቚክካኝቼቴቢኚውቪቤበቅቩቯቾቡሪሖቍኡያዝይድዳዲዛቆሃሗዎቖኑቁኰኔሂሇሩሒቬቂቦቀቧቨቿሢቕቛኩ ኡ ከኘኜቝኞቜኦሼሽሟቘ቞ሯሦኢሻሱሏወዋቋ�ገግዓቓየዉ�
openai/whisper-large-v3 (pseudo-label, language set)
null
ar
0.45361
0.0224
0.06537
openai/whisper-large-v3 via faster-whisper
wrong_language
cmn_Hani
0.0593
false
false
false
1
36.88
amh_u001177
1a_lessons
amh
train
FSI
17b.mp3
null
null
PD-US
17 U.S.C. § 105
1.66
18.1
1
337.15
338.81
ስስ ሰሰ ሲሳሲ ርር ሳ ምሚልም መሻመ ሚ ማማ ል ለረለ ረ ሮሮ ሡራሜሽሡ ሁሩህሱሁ ላሪሆሎላ
openai/whisper-large-v3 (pseudo-label, language set)
null
ur
0.53809
0.06226
0.00002
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.9009
false
true
false
1
34.34
amh_u002050
1a_lessons
amh
train
FSI
23b.mp3
null
null
PD-US
17 U.S.C. § 105
2.83
20.1
0.894
82.15
84.98
ስስ ሰሰ ርሳር ሴሲሚልለምመሎማሁህላሜሠሉራሕሊረሮሡሆሃሱሄሬሓሖሔሀሗሒሪሂሟሢሌሣሩሑሙሯሦሞሽሥሴ ሲ ም ማ ለ መ ሚ ረ ሮ ሠ ሜ ሙ ሖ ሪ ል ሢ ሡ ህ ሁ ላ ሣ ሉ ሀ ሳ ሞ ሩ ሃ ሎ ሟ ሗ ሆ ሂ ራ ሕ ሊ ሥ ሓ ሇ ሻ ሑ ሒ ሬ ሄ ሱ ሏ ሌ ሔ ሦ ሼ ሽ ሯ ሤ ሐሻሼሏሇሹሾሿሤሐ
openai/whisper-large-v3 (pseudo-label, language set)
null
cs
0.35596
0.15186
0.00022
openai/whisper-large-v3 via faster-whisper
unverified
tig_Ethi
0.6777
false
true
false
1
60.07
End of preview. Expand in Data Studio

AfriBase-Speech v1.0

Speech data for African languages built only from audio that is in the public domain in the United States (17 U.S.C. § 105, works of the US federal government). Version 1.0 is the release described in two papers at DHASA 2026:

  • Tier 1a, lesson audio. Language course recordings of the US Foreign Service Institute (FSI) for the paper's five evaluation languages, split into utterances, labelled with Whisper large-v3, with the English speech of the course instructor removed by language identification. Paper: AfriBase-Speech: Auditing, Tiering, and Packaging African Speech Resources from Government Public Domain Archives.
  • Tier 1b, Voice of America (VOA) broadcast audio. Segments of audio from VOA's African language services (plus Portuguese and Haitian Creole), crawled after the March 2025 shutdown of the language services. Paper: Archival ASR for African Broadcasting: Automatic Transcription of VOA African-Language Audio.

Maintainer: Ernst van Gassen, Independent researcher at Arktos AI Lab, ORCID 0009-0000-0314-6441. Pipeline code (MIT): https://github.com/arktos-ai-lab/afribase-speech-pipeline. Fine-tuned Whisper-small models from the AfriBase-Speech paper: vGassen/afribase-speech-whisper-small. Everything together: the AfriBase-Speech collection.

Contents

Every row is one utterance: the original 16 kHz mono WAV audio plus provenance metadata. Each tier and language is its own configuration.

Tier 1a, lesson audio (segments kept after English removal; 95/5 train/validation split as in the paper)

Config Language Segments Hours
lessons_amh Amharic 8,225 5.23
lessons_hau Hausa 914 0.38
lessons_lin Lingala 3,359 1.69
lessons_swh Swahili 3,656 1.57
lessons_yor Yoruba 2,510 1.40
Total 18,664 10.27

Tier 1b, VOA broadcast audio (one train split per language)

Config Service Source files Segments Hours Transcripts
voa_hau Hausa 1,303 191,302 220.29 fine-tuned Whisper-small, grade noisy
voa_por Portuguese (Africa) 2,328 116,773 235.73 Whisper large-v3, grade approximate
voa_hat Haitian Creole 1,607 55,680 122.36 Whisper large-v3, grade approximate
voa_lin Lingala 26 751 0.96 fine-tuned Whisper-small, grade noisy
voa_swh Swahili 6 134 1.35 fine-tuned Whisper-small, grade noisy
voa_amh Amharic 3 122 1.27 fine-tuned Whisper-small, grade noisy
voa_som Somali 1 24 0.24 Whisper large-v3, grade uninformative
voa_sna Shona 5 29 0.25 Whisper large-v3, grade uninformative
voa_nde Ndebele service 2,704 663,996 564.79 none (no model covers Ndebele)
voa_orm Afaan Oromo 120 19,465 17.26 Omnilingual ASR 7B (not graded; see per-row flags)
voa_tir Tigrinya 120 26,746 21.09 Omnilingual ASR 7B (not graded; see per-row flags)
voa_kin Kinyarwanda/Kirundi 120 48,509 40.88 Omnilingual ASR 7B (not graded; see per-row flags)
voa_bam Bambara 118 15,046 14.22 Omnilingual ASR 7B (not graded; see per-row flags)
Total 1,138,577 1,240.7

"Source files" counts the VOA audio or video files that yielded at least one segment. For Afaan Oromo, Tigrinya, Kinyarwanda/Kirundi and Bambara the VOA paper used a crawl capped at 120 media files per service.

Transcripts

All text in this release is machine output, never a human transcription. The text_source field names the model for every row.

  • Tier 1a labels come from Whisper large-v3 with the target language set in advance.
  • Tier 1b transcripts come from Whisper large-v3 or from Whisper-small fine-tuned on Tier 1a, as in the VOA paper. transcript_grade repeats the paper's per-language verdict: approximate (usable as approximate search text), noisy (a noisy index for exploration), uninformative (carries no information about the audio; kept for completeness, do not train on it).
  • Afaan Oromo, Tigrinya, Kinyarwanda/Kirundi and Bambara: Omnilingual ASR instead of MMS. The VOA paper reports transcripts from Meta's MMS model (mms-1b-all), whose weights are licensed CC BY-NC 4.0, which is not compatible with releasing this dataset under CC BY 4.0. Those transcripts are not included. Instead every segment of these four services was transcribed with Meta's Omnilingual ASR (omniASR_LLM_7B, Apache-2.0) with the language set in advance. The VOA paper's Afaan Oromo word error rate (63.9% on FLEURS) describes MMS; Omnilingual scores 55.0% on the same test set (see Differences from the papers). FLEURS is read speech, so neither figure describes broadcast audio directly. The other three languages have no FLEURS test set. transcript_grade is empty for these languages because the paper did not grade Omnilingual output; use the per-row flags.

Per-row quality flags

Automatic transcripts of low-resource languages contain hallucinations: text in an unrelated language (for example Turkish subtitle credits on Somali audio), looping output, or plausible-looking strings that are not words. Every transcribed row therefore carries reference-free checks, so you can filter rows instead of trusting a per-language average:

  • text_lid, text_lid_prob: the top label of the GlotLID text language identifier and its probability.
  • text_lid_match, text_script_match: whether that label is the target language, and whether it has the target's script.
  • text_repeated_trigram, text_dup_count: looping output within the transcript, and how often the identical text occurs in the same tier and language.
  • text_chars_per_s: characters per second of audio; extreme values suggest hallucination.
  • text_flag, a summary, applied in this order: empty; wrong_language (wrong script, or GlotLID names a specific other language with probability at least 0.5); repetitive (repeated trigram, or the same text 5 or more times); plausible (GlotLID confirms the target language); unverified (right script, but GlotLID does not confirm the language). For Amharic and Tigrinya, other Ethiopic-script labels count as unverified, because GlotLID cannot reliably separate Amharic, Tigre and Tigrinya on noisy text.

plausible is not a claim that a transcript is correct: GlotLID checks the language of the text, not whether it matches the audio. For training, a conservative choice is text_flag == "plausible" within the languages graded approximate. Rows without a transcript have null flags.

Config plausible unverified repetitive wrong_language
lessons_amh 142 7,145 31 907
lessons_hau 208 613 0 93
lessons_lin 1,572 1,611 0 176
lessons_swh 2,924 594 6 132
lessons_yor 99 1,893 5 513
voa_hau 59,097 114,875 1,169 16,161
voa_por 108,082 797 6,977 917
voa_hat 37,990 6,699 2,454 8,535
voa_lin 173 309 0 269
voa_swh 7 0 126 1
voa_amh 0 111 9 2
voa_som 0 0 0 24
voa_sna 0 12 5 12
voa_orm (Omnilingual) 16,396 1,955 253 861
voa_tir (Omnilingual) 14,140 6,171 428 6,007
voa_kin (Omnilingual) 43,504 2,074 1,648 1,283
voa_bam (Omnilingual) 6,748 7,034 100 1,164

(voa_hat also has 2 empty transcripts.) For voa_tir, most wrong_language rows are in Arabic script; the Tigrinya service's audio appears to include Arabic-language material, and these rows should not be used as Tigrinya. For voa_bam, GlotLID's Dyula label (dyu_Latn, a closely related Manding language) counts as unverified, not wrong_language. For voa_kin, Kirundi (run_Latn) counts as a match, because the service broadcasts in both. Yoruba lesson labels are mostly unverified because they lack the tone marks and subdots of standard orthography, which GlotLID relies on.

Differences from the papers

This release follows the two DHASA 2026 papers. Where it differs, the reason is the same in every case: the papers used Meta's MMS model (mms-1b-all), whose weights carry a non-commercial licence (CC BY-NC 4.0). Text produced with it cannot be released under CC BY 4.0, so this release uses Meta's Omnilingual ASR (omniASR_LLM_7B, Apache-2.0) instead.

What Paper says This dataset has Why
Transcripts for Afaan Oromo, Tigrinya, Kinyarwanda/Kirundi, Bambara MMS mms-1b-all Omnilingual ASR omniASR_LLM_7B, language set in advance MMS weights are CC BY-NC 4.0; Omnilingual is Apache-2.0
How many segments of those four languages are transcribed A random sample of 3,000 per language (to measure quality) Every segment: 109,766 A release should transcribe all of it; the sample was only a measurement
Word error rate on FLEURS (test split), Afaan Oromo MMS: 63.9% (CER 17.0%) Omnilingual: 55.0% (CER 14.4%), same 41 utterances and same text normalisation Different model; re-measured for this release
Word error rate, Tigrinya, Kinyarwanda/Kirundi, Bambara None (no FLEURS test set) None (no FLEURS test set) No change
transcript_grade for those four languages Graded from MMS output Empty The paper's grades describe MMS output; use the per-row flags instead
Per-row quality checks Per-language aggregates only text_flag and its evidence fields on every transcript Lets users drop unreliable rows (added for this release)
Ndebele service Segmented, not transcribed Same: no transcripts Omnilingual has no Ndebele model; MMS output is not releasable

Everything else (segments, Tier 1a labels from Whisper large-v3, Tier 1b Whisper transcripts and their grades, the 33 withheld Hausa segments, the train/validation split) is as in the papers.

Fields

Field Meaning
audio 16 kHz mono WAV
utterance_id stable segment ID
tier 1a_lessons or 1b_voa
language ISO 639-3 code of the course or the VOA service (not verified per segment, see Limitations)
split train or validation
source_type FSI (all Tier 1a segments in v1.0) or VOA_broadcast
source_file file the segment was cut from
source_url, article_id VOA page the media came from (Tier 1b, where recorded)
license, license_basis status of the source recording: PD-US, 17 U.S.C. § 105
duration_s, span_start_s, span_end_s segment length and position in the source file
snr_db, speech_fraction quality filter values
text, text_source, transcript_grade machine transcript, the model that made it, and the paper's grade
lid_top, lid_top_prob, lid_p_en, lid_p_target, lid_source Tier 1a only: Whisper large-v3 audio language identification used to remove English
text_flag, text_lid, text_lid_prob, text_lid_match, text_script_match, text_repeated_trigram, text_dup_count, text_chars_per_s per-row transcript quality checks (see Per-row quality flags)

Usage

from datasets import load_dataset

lessons = load_dataset("vGassen/afribase-speech", "lessons_swh", split="train")
voa = load_dataset("vGassen/afribase-speech", "voa_hau", split="train", streaming=True)

To download one language only: hf download vGassen/afribase-speech --repo-type dataset --include "voa/lin/*".

Datasheet

Tier 1a, lesson audio Tier 1b, VOA broadcast audio
Collection FSI course MP3s (the paper inventories FSI, Peace Corps and DLI audio for 22 African languages and Haitian Creole; v1.0 packages the five evaluation languages, all FSI); split at pauses (at least 500 ms below -35 dBFS), kept if 1 to 15 s, SNR at least 10 dB, speech fraction at least 0.60 Media from VOA language service sites (sitemaps and a legacy CMS ID scan), checked for page language; short clips split at 600 ms below -40 dBFS and kept if 1 to 30 s, SNR at least 6 dB and mostly speech; full broadcasts split at 200 ms below -38 dBFS without the SNR floor
Speakers One to three per course, chosen for clear pronunciation; slow, pedagogical speech Many speakers, news pace, radio compression, varying noise
Domain Greetings, travel, government services, basic commerce News and current affairs
Intended uses ASR fine-tuning and evaluation research for low-resource African languages Self-supervised pretraining, domain adaptation, search over the archive
Out of scope Speaker identification or voice cloning of the course speakers; treating labels as ground truth The same, and any use of non-VOA content inside broadcasts (see below)

Limitations

  • Labels and transcripts are machine output; their quality is bounded by the model's ability in each language.
  • Tier 1a removes English instructor speech only; the detector seldom names the target language itself, so other stray speech can remain. Like most kept Yoruba labels, Yoruba text lacks tone marks and subdots.
  • Tier 1b language labels come from the VOA service and page, not from the audio of each segment. Broadcasts switch to English or French, and the shared legacy ID space can let in pages of other services. In particular, a later audio language identification run (not part of v1.0) suggests that a large share of the voa_nde segments from the Ndebele service is in Shona.
  • Somali, Shona, Swahili and Amharic yield few Tier 1b segments because most of their files are long continuous broadcasts that the silence splitter could not cut. A later version will re-segment them.
  • 33 Hausa segments from three misattributed source files were withheld, as stated in the VOA paper.

Licence

Two layers. The source recordings are works of the US federal government and are in the public domain in the United States (17 U.S.C. § 105); the license field records this per row. Their status in other countries can differ. The compiled dataset (segmentation, filtering, labels, transcripts, metadata, packaging) is released under CC BY 4.0. VOA broadcasts can contain material VOA does not own, such as music, wire service clips or outside interviews; section 105 does not cover such material, and users are responsible for how they use it.

Citation

@misc{vangassen2026afribasespeechdata,
  author    = {van Gassen, Ernst},
  title     = {{AfriBase-Speech} (v1.0)},
  year      = {2026},
  publisher = {Hugging Face},
  doi       = {10.57967/hf/11727},
  url       = {https://huggingface.co/datasets/vGassen/afribase-speech}
}

@inproceedings{vangassen2026afribasespeech,
  author    = {van Gassen, Ernst},
  title     = {{AfriBase-Speech}: Auditing, Tiering, and Packaging African Speech Resources from Government Public Domain Archives},
  booktitle = {Proceedings of the Digital Humanities Association of Southern Africa Conference (DHASA 2026)},
  year      = {2026}
}

@inproceedings{vangassen2026voa,
  author    = {van Gassen, Ernst},
  title     = {Archival {ASR} for African Broadcasting: Automatic Transcription of {VOA} African-Language Audio},
  booktitle = {Proceedings of the Digital Humanities Association of Southern Africa Conference (DHASA 2026)},
  year      = {2026}
}
Downloads last month
61

Models trained or fine-tuned on vGassen/afribase-speech

Collection including vGassen/afribase-speech