01 · EVALUATE
Representative speech materials
Review Yorùbá audio and matching transcript materials before progressing to a commercial discussion.
Available to evaluate · Yorùbá speech
NDPC REGISTERED · NDPC/DCP/13596
Registered with the Nigeria Data Protection Commission. Certificate available on request. See our Compliance page and Privacy Policy.
Yorùbá speech data for AI training, with tonal accuracy built in.
Evaluate representative Yorùbá speech samples with time-aligned transcripts, or commission a scope built around your model, speakers, scenarios and delivery requirements. BSG DataWorks combines native-speaker language oversight, full tonal orthography, consent documentation and commercial licensing support.
Availability
Start with representative Yorùbá speech, then scope what your model needs next.
BSG DataWorks has Yorùbá speech and time-aligned transcription support available for technical evaluation. For production requirements, we will discuss the relevant usage rights, the current licensable scope and whether a commissioned collection or annotation extension is appropriate.
02 · SPECIFY
Language-aware scope
Describe the intended model use, speaker profile, scenario, volume, format, tonal-orthography requirements and timeline.
03 · LICENSE
Documented commercial use
Discuss the required licence, consent documentation, delivery materials and permitted commercial use as part of the project scope.
AI use cases
Build for Yorùbá as it is written and spoken.
For voice and language systems, the data brief should account for the target interaction, tonal orthography, speakers, recording environments, domains and evaluation criteria rather than treating Yorùbá as a generic-text or generic-speech case.
ASR and speech-to-text
Plan speech and time-aligned transcript requirements for systems that need Yorùbá language handling and appropriate written tone marks.
TTS and voice experiences
Discuss voices, recording requirements and contributor-rights scope for speech synthesis or branded voice work, subject to the agreed consent and licence terms.
Voice assistants and localisation
Define prompts, dialogue style, domains and speaker mix for Yorùbá-facing customer-service, information and voice-interface experiences.
Language-model evaluation
Use a carefully scoped sample or commissioned dataset to assess transcript treatment, language coverage or model behaviour against relevant Yorùbá input.
Technical and language requirements
Agree the language treatment and deliverables before collection starts.
Each production scope is documented around audio format, transcript treatment, tonal orthography, metadata, quality checks and delivery method. The exact package is agreed in the project statement of work or licence.
Typical audio and transcript options
- WAV audio, 16 kHz mono 16-bit PCM; 48 kHz studio option where agreed
- Verbatim and time-aligned transcript options
- Yorùbá transcripts with full tonal diacritics and Unicode NFC normalisation
- JSON, TXT or CSV delivery formats to brief
Quality and documentation
- Native-speaker review of 100% of Yorùbá transcriptions
- Inter-annotator agreement of ≥ 80% (word-level or Cohen’s κ per project), measured on a 10% blind sample per batch and reported in the batch QC report
- Free re-delivery on any batch failing agreed QA thresholds (up to two re-deliveries per batch)
- Audio screening for signal quality, clipping and duplication where relevant to the agreed delivery
- Data-card, licence and consent documentation materials for the delivery
From evaluation to delivery
A clear route from first sample to language-aware production scope.
Start with enough information for BSG DataWorks to assess fit, then agree the language, commercial and operational requirements before work begins.
Review
Evaluate representative material and identify whether an existing licence, custom extension or new collection discussion is appropriate.
Brief
Share the model use, target speakers, speech scenario, technical format, transcript treatment, required volume, timeline and governance needs.
Agree
Confirm delivery criteria, tonal-orthography treatment, rights, documentation and the commercial route in the applicable proposal, statement of work or licence.
Deliver
Receive the agreed audio, transcripts, metadata and documentation through client-specified secure transfer or an agreed approved destination.
Frequently asked questions
Questions before evaluating Yorùbá speech data.
Is Yorùbá speech data available to evaluate now?
BSG DataWorks has Yorùbá speech and time-aligned transcription support available for technical evaluation. Contact us with your intended use so we can discuss the relevant materials and the appropriate commercial path.
Can transcripts include tonal diacritics?
Yes. Yorùbá transcripts include full tonal diacritics and Unicode NFC normalisation by default. State any additional transcription, time-alignment, annotation and delivery-format requirements in your initial brief.
Can we commission a Yorùbá collection to our specification?
Yes. A commissioned discussion can cover the speech scenario, speaker profile, recording environment, target volume, transcript and metadata requirements, delivery timeline, rights and quality criteria.
How do commercial rights and consent documentation work?
Commercial use, permitted sublicensing and the documentation package are addressed in the applicable licence or project agreement. BSG DataWorks collects and maintains contributor-consent records consistent with the agreed scope.
Start a Yorùbá speech brief
Tell us what your model needs.
Share a few project details and we will reply within 48 hours with initial scoping questions, a likely timeline and the next step.
What helps us get started
- Intended model use and Yorùbá speech scenario
- Target speaker, market or participant profile
- Target volume, audio format and transcript requirements
- Tonal-orthography, annotation or metadata requirements
- Desired delivery timeline and governance requirements
- Approximate pilot budget or licensing range, if known
If you do not have all of this yet, send what you have — including if you are still scoping the budget. We will fill in the rest on a call.
Prefer to write directly? dataservices@bsgdataworks.com
WhatsApp only: +234 806 790 4903