Ctrl+K
Ctrl+K

Bulgarian CV dictionary v2.0.0#

  • Maintainer: Vox Communis

  • Language: Bulgarian

  • Dialect: N/A

  • Phone set: XPF

  • Number of words: 9,757

  • Phones: a aː b d dz dʒ dː e eː f i iː j k m n nː o oː p r s t ts tsː tʃ tː u uː v x z zː ɡ ɤ ɫ ɫː ʃ ʒ

  • License: CC-0

  • Compatible MFA version: v2.0.0

  • Citation:

@misc{Ahn_Chodroff_2022,
	author={Ahn, Emily and Chodroff, Eleanor},
	title={VoxCommunis Corpus},
	address={\url{https://osf.io/t957v}},
	publisher={OSF},
	year={2022},
	month={Jan}
}

Acoustic models

Installation#

Install from the MFA command line:

mfa model download dictionary bulgarian_cv

Or download from the release page.

Intended use#

This dictionary is intended for forced alignment of Bulgarian transcripts.

This dictionary uses the XPF phone set for Bulgarian, and was used in training the Bulgarian XPF acoustic model. Pronunciations can be added on top of the dictionary, as long as no additional phones are introduced.

Performance Factors#

When trying to get better alignment accuracy, adding pronunciations is generally helpful, especially for different styles and dialects. The most impactful improvements will generally be seen when adding reduced variants that involve deleting segments/syllables common in spontaneous speech. Alignment must include all phones specified in the pronunciation of a word, and each phone has a minimum duration (by default 10ms). If a speaker pronounces a multisyllabic word with just a single syllable, it can be hard for MFA to fit all the segments in, so it will lead to alignment errors on adjacent words as well.

Ethical considerations#

Deploying any Speech-to-Text model into any production setting has ethical implications. You should consider these implications before use.

Demographic Bias#

You should assume every machine learning model has demographic bias unless proven otherwise. For pronunciation dictionaries, it is often the case that transcription accuracy and lexicon coverage for the prestige variety modeled in this dictionary compared to other variants. If you are using this dictionary in production, you should acknowledge this as a potential issue.

IPA Charts#

Consonants#

Obstruent symbols to the left of are unvoiced and those to the right are voiced.

Manner

Labial

Labiodental

Alveolar

Alveopalatal

Palatal

Velar

Nasal
Occurrences:
1,667
Examples:
* много:
[m n o ɡ o]
* милин:
[m i ɫ i n]
* молба:
[m o ɫ b a]
* мария:
[m a r i j a]
Occurrences:
4,262
Examples:
* много:
[m n o ɡ o]
* пенев:
[p e n e v]
* милин:
[m i ɫ i n]
* любен:
[ɫ j u b e n]
Occurrences:
42
Examples:
* ценно:
[ts e nː o]

Stop
Occurrences:
2,437
Examples:
* пенев:
[p e n e v]
* плет:
[p ɫ e t]
* топка:
[t o p k a]
* топът:
[t o p ɤ t]
Occurrences:
1,273
Examples:
* гръб:
[ɡ r ɤ b]
* обзе:
[o b z e]
* молба:
[m o ɫ b a]
* любен:
[ɫ j u b e n]
Occurrences:
4,835
Examples:
* часът:
[tʃ a s ɤ t]
* живот:
[ʒ i v o t]
* усети:
[u s e t i]
* акт:
[a k t]
Occurrences:
42
Examples:
* оттук:
[o tː u k]
* оттам:
[o tː a m]
Occurrences:
2,036
Examples:
* идеха:
[i d e x a]
* видял:
[v i d j a ɫ]
* осъди:
[o s ɤ d i]
* думай:
[d u m a j]
Occurrences:
1
Examples:
Occurrences:
2,350
Examples:
* ярко:
[j a r k o]
* ако:
[a k o]
* чайки:
[tʃ a j k i]
* моско:
[m o s k o]
Occurrences:
1,114
Examples:
* гръб:
[ɡ r ɤ b]
* много:
[m n o ɡ o]
* фланг:
[f ɫ a n ɡ]
* негов:
[n e ɡ o v]

Affricate
Occurrences:
524
Examples:
* целия:
[ts e ɫ i j a]
* ръце:
[r ɤ ts e]
* вълци:
[v ɤ ɫ ts i]
* цял:
[ts j a ɫ]
Occurrences:
1
Examples:
Occurrences:
5
Examples:
Occurrences:
1,347
Examples:
* чайки:
[tʃ a j k i]
* число:
[tʃ i s ɫ o]
* часът:
[tʃ a s ɤ t]
* лично:
[ɫ i tʃ n o]
Occurrences:
49
Examples:
* хаджи:
[x a dʒ i]
* ходжа:
[x o dʒ a]
* джубе:
[dʒ u b e]
* джеба:
[dʒ e b a]

Sibilant
Occurrences:
2,833
Examples:
* осъди:
[o s ɤ d i]
* число:
[tʃ i s ɫ o]
* часът:
[tʃ a s ɤ t]
* усети:
[u s e t i]
Occurrences:
1,700
Examples:
* обзе:
[o b z e]
* без:
[b e z]
* юзда:
[j u z d a]
* знаех:
[z n a e x]
Occurrences:
4
Examples:
* иззад:
[i zː a d]
Occurrences:
1,150
Examples:
* наши:
[n a ʃ i]
* пише:
[p i ʃ e]
* пеш:
[p e ʃ]
* лижеш:
[ɫ i ʒ e ʃ]
Occurrences:
574
Examples:
* живот:
[ʒ i v o t]
* тежко:
[t e ʒ k o]
* лижеш:
[ɫ i ʒ e ʃ]
* грижи:
[ɡ r i ʒ i]

Fricative
Occurrences:
151
Examples:
* фланг:
[f ɫ a n ɡ]
* филип:
[f i ɫ i p]
* чифт:
[tʃ i f t]
* анифе:
[a n i f e]
Occurrences:
3,013
Examples:
* пенев:
[p e n e v]
* видял:
[v i d j a ɫ]
* живот:
[ʒ i v o t]
* две:
[d v e]

Approximant
Occurrences:
1,913
Examples:
* ярко:
[j a r k o]
* видял:
[v i d j a ɫ]
* любен:
[ɫ j u b e n]
* чайки:
[tʃ a j k i]

Trill
Occurrences:
3,850
Examples:
* ярко:
[j a r k o]
* гръб:
[ɡ r ɤ b]
* орел:
[o r e ɫ]
* мария:
[m a r i j a]

Lateral
Occurrences:
3,081
Examples:
* видял:
[v i d j a ɫ]
* милин:
[m i ɫ i n]
* молба:
[m o ɫ b a]
* орел:
[o r e ɫ]
Occurrences:
4
Examples:
* селлм:
[s e ɫː m]
* аллах:
[a ɫː a x]

Vowels#

Vowel symbols to the left of are unrounded and those to the right are rounded.

Front

Near-Front

Central

Near-Back

Back

Close
Occurrences:
5,760
Examples:
* идеха:
[i d e x a]
* видял:
[v i d j a ɫ]
* милин:
[m i ɫ i n]
* осъди:
[o s ɤ d i]
Occurrences:
24
Examples:
* чиито:
[tʃ iː t o]
Occurrences:
1,461
Examples:
* любен:
[ɫ j u b e n]
* думай:
[d u m a j]
* усети:
[u s e t i]
* плюна:
[p ɫ j u n a]
Occurrences:
1
Examples:

Close-Mid
Occurrences:
5,835
Examples:
* идеха:
[i d e x a]
* пенев:
[p e n e v]
* обзе:
[o b z e]
* орел:
[o r e ɫ]
Occurrences:
30
Examples:
* пееше:
[p eː ʃ e]
* живее:
[ʒ i v eː]
* смее:
[s m eː]
* пееха:
[p eː x a]
Occurrences:
1,525
Examples:
* гръб:
[ɡ r ɤ b]
* осъди:
[o s ɤ d i]
* часът:
[tʃ a s ɤ t]
* ръце:
[r ɤ ts e]
Occurrences:
5,451
Examples:
* ярко:
[j a r k o]
* много:
[m n o ɡ o]
* обзе:
[o b z e]
* молба:
[m o ɫ b a]
Occurrences:
12
Examples:

Open-Mid

Open
Occurrences:
10,283
Examples:
* идеха:
[i d e x a]
* ярко:
[j a r k o]
* видял:
[v i d j a ɫ]
* молба:
[m o ɫ b a]
Occurrences:
4
Examples:
* а-а:
[aː]
On this page
Show Source