Transcription apps for Xitsonga: xiRonga speech

Munghana Lonene FM, the SABC's Xitsonga station: transcription apps compared for Xitsonga, in South Africa, Mozambique and Zimbabwe.

This page compares transcription apps and services for Xitsonga.

Speakers. About 3.7 million native speakers (2006 to 2011, Ethnologue 18th ed.) and about 3.4 million second-language speakers in South Africa (2002, Webb). Both figures are old.

Writing. Latin. The source does not mention tone marks.

Where Xitsonga is recorded

  • Radio. Munghana Lonene FM, the SABC's Xitsonga station, began on 1 February 1965 as Tsonga programmes from the Johannesburg studios (later Radio Tsonga).
  • Radio (Munghana Lonene FM). Xitsonga radio on Munghana Lonene FM uses English insertions that listeners notice.
  • Classroom. A University of Limpopo MA study (Baloyi 2010) says Xitsonga-English switching is common among high school learners, teachers and university students; the study itself examined Xitsonga literature and textbooks.

Xitsonga varieties a tool has to cope with

  • Tswa (Khambana) and Ronga. Mozambique. Mozambique recognises the Tswa-Ronga languages as minority languages. The source's spelling and pronunciation vary across dialects.

Spelling, tone and names in Xitsonga

The language has several names, so a search or transcript may use Xitsonga, Tsonga, Thonga, Changana, Shangana or Shangani.

Where Xitsonga is spoken

  • South Africa. Official language under the 1996 Constitution; spoken in Limpopo, Mpumalanga, Gauteng, North West.
  • Mozambique. Recognised minority language (Changana/Xichangana); Maputo, Gaza and Inhambane provinces are among the regions listed.
  • Zimbabwe. Official under the name Shangani; Chiredzi and Mwenezi are among the districts listed.
  • Eswatini. Listed as a country with speakers; not official in Eswatini.

Transcription apps compared for Xitsonga

OptionWhat its page shows about XitsongaSource
AutransAutrans offers Xitsonga. Measured result: Word error rate 71.1% on read speech, NCHLT test set, measured Sep 2026.autrans.online
SonixThe list on the Sonix page does not include Xitsonga.sonix.ai
DescriptThe list on the Descript page does not include Xitsonga.help.descript.com
TrintThe list on the Trint page does not include Xitsonga.info.trint.com
MaestraThe list on the Maestra page does not include Xitsonga.maestra.ai
VEEDVEED names Xitsonga on its page. It says: "Transcription: Afrikaans, Amharic, Arabic, Hausa, Igbo, Lingala, Shona, Somali, Swahili, Tsonga, Tswana, Xhosa, Yoruba, Zulu."support.veed.io
TurboScribeThe pages read for TurboScribe say nothing about Xitsonga.turboscribe.ai
RevThe pages read for Rev say nothing about Xitsonga.rev.com
Happy ScribeThe pages read for Happy Scribe say nothing about Xitsonga.happyscribe.com

Named: its page names the language. Not listed: its own list leaves the language out. Not stated: the pages read say nothing either way, so test a file. All pages read on 22 October 2026.

Speech APIs for developers: Xitsonga

APIWhat its page shows about XitsongaSource
SpeechmaticsThe list on the Speechmatics page does not include Xitsonga.docs.speechmatics.com
IntronThe list on the Intron page does not include Xitsonga. It says: "Not present: Lingala, Somali, Tsonga, Venda. intron.io markets '57 languages' and '63 total'."docs.voice.intron.io
Lelapa AIThe list on the Lelapa AI page does not include Xitsonga. It says: "Xhosa, Tswana, Tsonga, Swahili, Shona, Hausa and Yoruba appear only for text translation, so they are 'no' for speech-to-text."docs.lelapa.ai
AssemblyAIThe pages read for AssemblyAI say nothing about Xitsonga.assemblyai.com
SpitchThe pages read for Spitch say nothing about Xitsonga.docs.spitch.app

These are APIs for developers, not apps for a recording.

Where another tool is the better choice

  • VEED names Xitsonga on its own page. If that vendor's other features fit your work better, test it with a Xitsonga file first.
  • For a record that must stand up in court, use a human transcriber who speaks Xitsonga.

What Autrans offers for Xitsonga

Word error rate 71.1% (95% confidence interval 66.9–75.2%) on read speech, NCHLT test set, measured Sep 2026. Measured on 250 recordings. Word error rate is the share of words that differ from a human-written transcript, so lower is better. Read speech means people reading set sentences aloud, so conversation, background noise and phone-quality audio will usually score worse. Test set: NCHLT speech corpus (SADiLaR).

Autrans turns audio, video, WhatsApp voice notes and phone call recordings into a transcript you can read, search, edit and export.

Languages here: Xitsonga.

Questions people ask

Does Autrans transcribe Xitsonga?

Yes, Xitsonga is one of the live languages. Word error rate 71.1% (95% confidence interval 66.9–75.2%) on read speech, NCHLT test set, measured Sep 2026.

Do other tools name Xitsonga?

VEED names Xitsonga on its own pages, as read on 22 October 2026. Sonix, Descript and Trint do not list it.

Related

Start transcribing free

Send Autrans into your next meeting, use it in the room, or upload a recording. Your first 30 minutes are free, given once, and no credit card is needed to start.

Last checked 22 October 2026. How we measure accuracy. Spotted a mistake? Write to hello@autrans.online.