Common Voice Spontaneous Speech 4.0 - Sena

seh — Sena (seh)

This datasheet is for sps-corpus-4.0-2026-06-12 of the Mozilla Common Voice Spontaneous Speech dataset for Sena [seh - seh]. The dataset contains 290 clips representing 1.62 hours of recorded speech (0 hours validated) from 5 speakers.

Language

The Sena language (hereinafter Cisena, N 44 in the Guthrie classification, 1967-71), is a language originally belonging to the Congo-Kordophonian (Niger-Congo) family of the Bantu group. In Mozambique the Sena language is spoken predominantly in the provinces of lower Zambeze, namely Sofala, Zambezia, Tete and Manica and in the far south of the Republic of Malawi. The record of the presence of Sena speakers in the region is documented with the designation of the Village of Sena, from 1561. In Mozambique the Sena language is spoken to about 1,551,684 speakers of five or more years of age (INE, 2017) and in the Republic of Malawi it is spoken to about half a million people. The Cisena has the following variants: Sena Tonga, spoken in the north and center of Sofala, and in the borders of Tete and Zambezia; Sena Bangwe, spoken in the Beira area; Sena Phodzo, spoken between Sofala and Zambezia (from Marromeu, until Chinde) and Mopeia (in Zambezia); Sena Gombe, spoken in Caia, Mutarara, Chemba (coast), and Cheringoma and the coastal part of Zambezia; Sena Gorongozi, spoken in the Sierra Gorongosa area. The first grammar of the sena language was elaborated and printed in Chupanga (Marromeu), in 1900, by the french jesuit missionary and philologist Jules Torrend, in a trilingual edition in portuguese, sena and english. Two years earlier, the typography of this catholic mission had already produced a sena-portuguese catechism. In 1957, the first Portuguese-Sena-Portuguese dictionary was published in Beira. Since 2004, in Mozambique, the sena language has been part of bilingual education in rural schools in Sofala province, during the first five years of primary education, of the National Education System

Data splits for modelling

The dataset clips are categorised by transcription status and training-set assignment. The following tables summarise the distribution.

Audio clips

Training splits

Training split coverage: 0 of 1 transcribed & validated clips (0.0%)

Transcriptions

Transcription status

Writing system

The writing system uses the Latin script in general, but with some combinations of consonants that serve to represent the specific sounds of the language as ny, dz, jh, pf, bz, ng', ps, sv, ts, dh and bh.

Symbol table

Samples

Questions

There follows a randomly selected sample of questions used in the corpus.

1. Thangweranji azungu akhafuna tayi kuti anhtu apsipa ayenderere kutsogolo na mapfundziro makulumakulu?

2. Pisafunikana kuti anthu adye mwadidi mwene ntsomba, thangweranji? Towera akhale na thupi ya kutani?

3. Thangwiranji pasaperekwa luphato pangakhala pawoneka kusembana kwa anhtu?

4. Musaphika tani therere?

5. Pangakhala palibe kugumanika munthu wa kupita kufa omwe ndi wa dzindza, ule munthu wa pezi angapita kufa kutomera kwa ntsiku ineyo iye anacita khundu ya dzindza ya ale womwe adaferwa? Pisacitika tani?

Responses

There follows a randomly selected sample of transcribed responses from the corpus.

1. Ule ana uthambi ndi kulonga kuti iye ndi munthu ana uthambi kakamwe, asacita pinthu piyakunama ninga kunyengeza andzace na pinthu pinangopingango.

2. Dhuku kwa antu a cisena ndi cinthu comwe cisamangwa munsolo na azimayi na acikazi basi wokha.

3. Thangweyanji azungu akhafuna tayi kuti anthu apsipa ayenderere kutsogolo makumakulu? Iwo azungu akhakhondesa anthu kupfundza kuti ayende na mapfundziro makulumakulu kutsogolo na thangweranji? Na thangwera iwo akhagopa kuti pinango iye munthu svipa uyu angapfundza kakamwe, iye anatitengera mpando, anafuna kuti atenge dziko, iye anazindilira kuti dziko iyi njace natenepa anatitengera, anatithamangisa. Natenepa pisafunika kuti iye akhonde kuyenda kutsogolo na mapfundziro makulumakulu. Na thangwera iye khabe kudziwa dziko iyi njace. Ana ufulu , asafuna kukhala mwadidi m'dziko muno. Natenepa azungu penu atsamunda akhatikhondesa kuti ife tiyende na mapfundziro kutsogolo. Iwo akhacitisa kuti ife tipfundze mpaka kuyenda mapfundziro manayi okha basi. Mapfundziro anango akhatikhondesa kuti ife leka pfundze, mbatileke kupfundze kuti tiyende kutsogolo. Iwo akhatithamangisa tingafuna kuyenda kucita mapfundziro makulumakulu. Natenepa, pasalongwa kuti atsamunda akhadayipa moyo thangwera akhafuna tayi kuti iwo apfundzise anthu akafiri penu apsipa mapfundziro makulumakulu, thangwera angadazindikira kuti dziko iyi njawo na iwo asafunika kuti akhale ntendere.

Fields

Each row of a tsv file represents a single audio clip, and contains the following information:

• client_id - hashed UUID of a given user

• audio_id - numeric id for audio file

• audio_file - audio file name

• duration_ms - duration of audio in milliseconds

• prompt_id - numeric id for prompt

• prompt - question for user

• transcription - transcription of the audio response

• votes - number of people that who approved a given transcript

• age - age of the speaker1

• gender - gender of the speaker1

• language - language name

• split - for data modelling, which subset of the data does this clip pertain to

• char_per_sec - how many characters of transcription per second of audio

• quality_tags - some automated assessment of the transcription--audio pair, separated by |

  • transcription-length - character per second under 3 characters per second

  • speech-rate - characters per second over 30 characters per second

  • short-audio - audio length under 2 seconds

  • long-audio - audio length over 5 minutes

  • non-allowed-script - transcription contains characters from a writing system not associated with the language

  • mixed-script-words - a single word contains characters from multiple writing systems

  • mixed-script-transcription - transcription spans multiple writing systems, but each word consistently uses only one

Get involved

Community links

• Common Voice translators on Pontoon

• Common Voice Communities

Discussions

• Common Voice on Matrix

• Common Voice on Discourse

• Common Voice on Discord

• Common Voice on Telegram

Contribute

• Contribute questions

• Validate questions

• Answer questions

• Transcribe recordings

• Validate transcriptions

Acknowledgements

Datasheet authors

• Ricardo Samuel Bulaque <rbulaque@isced.ac.mz>

Licence

This dataset is released under the Creative Commons Zero (CC-0) licence. By downloading this data you agree to not determine the identity of speakers in the dataset.

Footnotes

1. For a full list of age, gender, and accent options, see the demographics spec. These will only be reported if the speaker opted in to provide that information. ↩ ↩2