Ctrl+K
Ctrl+K

Swedish CV dictionary v2.0.0#

  • Maintainer: Vox Communis

  • Language: Swedish

  • Dialect: N/A

  • Phone set: XPF

  • Number of words: 22,202

  • Phones: b d eː f g h iː j k l m n oː p q r s t uː v yː z øː ŋ œ ɐ ɑː ɔ ɕ ɛ ɛː ɡ ɧ ɪ ɵ ʉː ʏ

  • 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 swedish_cv

Or download from the release page.

Intended use#

This dictionary is intended for forced alignment of Swedish transcripts.

This dictionary uses the XPF phone set for Swedish, and was used in training the Swedish 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

Palatal

Velar

Uvular

Glottal

Nasal
Occurrences:
4,279
Examples:
* skam,:
[ɧ ɑː m]
* mats,:
[m ɐ t s]
* muta:
[m ʉː t ɑː]
* mitt:
[m ɪ t]
Occurrences:
12,324
Examples:
* älgen:
[ɛ j eː n]
* "din:
[d iː n]
* boden:
[b uː d eː n]
* giant:
[j iː ɐ n t]
Occurrences:
1,330
Examples:
* gång?:
[ɡ ɔ ŋ]
* längs:
[l ɛ ŋ s]
* äng:
[ɛ ŋ]
* trång:
[t r ɔ ŋ]

Stop
Occurrences:
2,716
Examples:
* parks:
[p ɐ r k s]
* uppe,:
[ɵ p eː]
* djupa:
[j ʉː p ɑː]
* post-:
[p ɔ s t]
Occurrences:
2,734
Examples:
* boden:
[b uː d eː n]
* begav:
[b eː ɡ ɑː v]
* britt:
[b r ɪ t]
* brev.:
[b r eː v]
Occurrences:
11,976
Examples:
* sätt?:
[s ɛ t]
* sålt.:
[s ɔ l t]
* mats,:
[m ɐ t s]
* muta:
[m ʉː t ɑː]
Occurrences:
6,545
Examples:
* ljud.:
[j ʉː d]
* jagad:
[j ɑː ɡ ɑː d]
* "din:
[d iː n]
* ord!:
[ɔ r d]
Occurrences:
4,695
Examples:
* sak!:
[s ɑː k]
* parks:
[p ɐ r k s]
* rik!:
[r iː k]
* verk.:
[v ɛ r k]
Occurrences:
3,223
Examples:
* jagad:
[j ɑː ɡ ɑː d]
* gång?:
[ɡ ɔ ŋ]
* vågat:
[v oː ɡ ɑː t]
* begav:
[b eː ɡ ɑː v]
Occurrences:
17
Examples:
* qvick:
[q v ɪ k]
* queen:
[q ʉː eː eː n]
* qing:
[q ɪ ŋ]

Sibilant
Occurrences:
9,987
Examples:
* sätt?:
[s ɛ t]
* sålt.:
[s ɔ l t]
* sak!:
[s ɑː k]
* parks:
[p ɐ r k s]
Occurrences:
4
Examples:
* jazz:
[j ɐ z]
* buzz:
[b ɵ z]
Occurrences:
1,482
Examples:
* kemi.:
[ɕ eː m iː]
* maken:
[m ɑː ɕ eː n]
* säker:
[s ɛː ɕ eː r]
* okej!:
[uː ɕ eː j]

Fricative
Occurrences:
3,131
Examples:
* ifall:
[iː f ɐ l]
* före:
[f øː r eː]
* f.d.:
[f d]
* färd:
[f ɛ r d]
Occurrences:
3,646
Examples:
* vill.:
[v ɪ l]
* verk.:
[v ɛ r k]
* två:
[t v oː]
* vitt.:
[v ɪ t]
Occurrences:
1,914
Examples:
* human:
[h ʉː m ɑː n]
* hel:
[h eː l]
* hamra:
[h ɐ m r ɑː]
* hör:
[h øː r]

Approximant
Occurrences:
2,571
Examples:
* älgen:
[ɛ j eː n]
* ljud.:
[j ʉː d]
* jagad:
[j ɑː ɡ ɑː d]
* giant:
[j iː ɐ n t]

Trill
Occurrences:
14,568
Examples:
* parks:
[p ɐ r k s]
* lära:
[l ɛː r ɑː]
* rik!:
[r iː k]
* ord!:
[ɔ r d]

Lateral
Occurrences:
7,685
Examples:
* sålt.:
[s ɔ l t]
* ifall:
[iː f ɐ l]
* lära:
[l ɛː r ɑː]
* vill.:
[v ɪ l]

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:
4,059
Examples:
* ifall:
[iː f ɐ l]
* rik!:
[r iː k]
* "din:
[d iː n]
* kemi.:
[ɕ eː m iː]
Occurrences:
642
Examples:
* wily:
[v iː l yː]
* blyg,:
[b l yː ɡ]
* york.:
[yː ɔ r k]
* hyser:
[h yː s eː r]
Occurrences:
1,372
Examples:
* ljud.:
[j ʉː d]
* muta:
[m ʉː t ɑː]
* human:
[h ʉː m ɑː n]
* djupa:
[j ʉː p ɑː]
Occurrences:
3,206
Examples:
* boden:
[b uː d eː n]
* nato:
[n ɑː t uː]
* rop.:
[r uː p]
* om!:
[uː m]
Occurrences:
4,352
Examples:
* mitt:
[m ɪ t]
* vill.:
[v ɪ l]
* vitt.:
[v ɪ t]
* mill:
[m ɪ l]
Occurrences:
777
Examples:
* synd,:
[s ʏ n d]
* ylva:
[ʏ l v ɑː]
* ryska:
[r ʏ ɧ ɑː]
* tytt:
[t ʏ t]

Close-Mid
Occurrences:
11,056
Examples:
* älgen:
[ɛ j eː n]
* kemi.:
[ɕ eː m iː]
* boden:
[b uː d eː n]
* före:
[f øː r eː]
Occurrences:
1,197
Examples:
* före:
[f øː r eː]
* över.:
[øː v eː r]
* hör:
[h øː r]
* höra.:
[h øː r ɑː]
Occurrences:
2,078
Examples:
* dumt!:
[d ɵ m t]
* uppe,:
[ɵ p eː]
* putte:
[p ɵ t eː]
* blund:
[b l ɵ n d]
Occurrences:
1,097
Examples:
* två:
[t v oː]
* vågat:
[v oː ɡ ɑː t]
* svår:
[s v oː r]
* vrån:
[v r oː n]

Open-Mid
Occurrences:
6,891
Examples:
* sätt?:
[s ɛ t]
* älgen:
[ɛ j eː n]
* verk.:
[v ɛ r k]
* text.:
[t ɛ k s t]
Occurrences:
819
Examples:
* lära:
[l ɛː r ɑː]
* väg-:
[v ɛː ɡ]
* säker:
[s ɛː ɕ eː r]
* länet:
[l ɛː n eː t]
Occurrences:
1,458
Examples:
* löpte:
[l œ p t eː]
* föll.:
[f œ l]
* börje:
[b œ r j eː]
* rört:
[r œ r t]
Occurrences:
3,496
Examples:
* sålt.:
[s ɔ l t]
* ord!:
[ɔ r d]
* gång?:
[ɡ ɔ ŋ]
* roll.:
[r ɔ l]
Occurrences:
5,478
Examples:
* parks:
[p ɐ r k s]
* ifall:
[iː f ɐ l]
* mats,:
[m ɐ t s]
* akka.:
[ɐ k ɑː]

Open
Occurrences:
10,206
Examples:
* skam,:
[ɧ ɑː m]
* sak!:
[s ɑː k]
* jagad:
[j ɑː ɡ ɑː d]
* lära:
[l ɛː r ɑː]
On this page
Show Source