Transcribe long Yoruba recordings, with an English line under each part
Upload a Yoruba recording, such as a sermon, a family gathering or an interview, and get the whole thing written out in Yoruba with timestamps and an English translation beneath every paragraph. A recording no longer than 15 minutes is free in full. For longer audio, the first part is free and the full transcript, up to 6 hours, is $5.
- Yoruba in Latin letters; tone marks and underdots may be missing or partial
- English inside Yoruba speech, common in Lagos, kept in English
- Speakers from Nigeria, Benin and the diaspora
- English translation preselected under every paragraph
- One payment of $5 for up to 6 hours
Real output from a demo recording. Click a timestamp to see where it sits in the audio.
Tone marks and underdots: what to expect
Yoruba is a tonal language. Careful written Yoruba marks high and low tones with accents and uses underdots for the letters ẹ, ọ and ṣ. Without those marks, the same letters can stand for several different words. Many Yoruba speakers type without marks every day, and the transcript can be the same: diacritics may be missing, partial or occasionally wrong.
For reading, sharing and searching, an unmarked or partly marked text is usually workable. If you need fully marked Yoruba, for a publication or for teaching, plan to add or correct the marks yourself, and check the free first part to see how your recording comes out.
Sermons, services and long gatherings
Some of the longest Yoruba recordings come from church services, vigils and Islamic lectures, where preaching moves between Yoruba and English and songs come in between. Speech is transcribed; singing over loud instruments is much harder and may be marked [inaudible]. Bible or Quran references are written as the preacher says them.
Weddings, naming ceremonies and burials produce long family videos too, with oriki and greetings that elders treasure. Praise poetry is dense and full of old words and names, so expect more mistakes in those passages than in ordinary talk.
Reading it abroad
Yoruba speakers live in southwest Nigeria, in Benin and Togo, and in large communities in the UK, the US and Canada. Children raised abroad often understand some Yoruba but cannot read it well. The English under each paragraph lets them follow a grandparent's recording, and the Yoruba stays above it as the record of what was actually said.
Good to know
- Yoruba has less training audio than English; judge the free first part before you pay
- Names and places are written as heard, often without tone marks
- Speaker labels are added when the change of speaker is clear, and can be renamed on screen
- The result is an AI draft, not a certified transcript or sworn translation
How the tool works, what it costs and how your files are handled is on the Long Audio & Video to Text page; prices for every tool are on the pricing page.
Last updated
Questions and answers
Will the Yoruba have all its tone marks?
Not reliably. Tone marks and underdots may be missing or partial, so add or correct them if you need fully marked text.
Why can't a chat assistant do my two-hour Yoruba service?
Audio uploads in chat assistants are limited in length, and a long transcript does not fit in one reply, so the result is shortened or stops. This tool transcribes the service in parts and joins them with timestamps.
How much is a long recording?
$5, once, for a recording up to 6 hours. Unlimited is $19 a week for recordings of up to 12 hours; it renews weekly and you can cancel any time.
Can I upload a video from a party or service?
Yes. MP4, MOV and other video files up to 2 GB are accepted, and only the sound is transcribed. YouTube links are not taken; download the file first.
Who can see my recording?
Only you. It is sent to an AI model provider only to be transcribed, never used to train models, and jobs without an account are deleted after 7 days.