Echo voices and Baobab national models — Africa-first language intelligence.

Explore Baobab
Skip to content
Open consoleLog in

Lugemi status

Baobab Yoruba model card

Nigeria regional Yoruba (Lagos / Oyo) national model under Baobab — spectral_gate_v2 clarity, national tiers, research slot, unlock gates.

Translate (en → Twi)Baobab canopyAll status pagesMachine JSON
model_id
baobab-yo
display_name
Baobab Yoruba
model_version
2026.10.national
family
baobab
status
national_shipped
product_line
National Lagos / Oyo Yoruba speech + phrase MT corridor — generative_inference_live false; neural_nmt false; trainingEligible locked. Product default yo-NG / own:yo-ng-male; Egba secondary.
as_of
2026-10-23T09:40:29.084Z

varieties

varieties: 1

  • Yoruba (Lagos / Oyo)

    variety_id
    yo-ng
    pack_key
    lagos
    name_en
    Yoruba (Lagos / Oyo)
    bcp47
    yo-NG
    iso6393
    yor
    echo_voice_id
    own:yo-ng-male
    mt_lang_key
    yo
    language_pack_version
    yo-ng-pack-2026.10
    corridor_id
    en-yo-lagos
    primary_use
    Lagos / Southwest Nigeria citizen services / education / MeetingBridge Nigeria Yoruba
    dataset_manifest
    apps/api/eval/corridors/en-yo-lagos/test-scripts.json
    trainingEligible
    false
    audio_clean
    spectral_gate_v2
    speech_finetune_recipe
    docs/baobab/SPEECH_FINETUNE_RECIPE.md

    english_accent

    locale
    en-NG
    voice_id
    own:en-ng-female
    label
    Chioma · Nigerian English
    region
    nigeria
    note
    Nigerian English accent paired with Nigeria L1 national models (Yoruba, Igbo, Hausa Kano).

    domains

    domains: 6
    • greetings
    • everyday
    • airtime
    • health
    • civic
    • market

    not_interchangeable_with

    not_interchangeable_with: 4
    • yo-ng-egba
    • yo-ng-ijebu
    • yo-bj-anago
    • ha-ng-kano

    mt

    status
    phrase_table_only
    note
    Deterministic phrase + lexical Baobab pack in lugemi-mt.engine. Fail-closed when uncovered. No dedicated neural NMT service.

    pairs

    pairs: 1
    • en↔yo

    asr

    status
    neural_offline
    checkpoint
    openai/whisper-small (Yoruba + Lagos phrase rescoring)
    license
    see_model_card
    env_gate
    Yoruba speech
    variety_exclusive_finetune
    false

    tts

    status
    neural_offline
    checkpoint
    facebook/mms-tts-yor ONNX + Lagos prosody
    license
    CC-BY-NC-4.0
    env_gate
    Yoruba speech
    variety_exclusive_finetune
    false

    s2s_corridor

    status
    corridor_pass
    note
    en-yo-lagos corridor scripts PASS when speech URL points at lugemi-yoruba. Broad domain deferred. Clarity: spectral_gate_v2.

national_tiers

national_tiers: 4

  • Public demo

    id
    public_demo
    name
    Public demo
    who
    Citizens, press, students (unauthenticated demos)
    auth
    None — soft abuse controls on demo speech
    maps_to_plan
    null
    residency_preference
    null
    audit
    false
    commercial_nc_clean_tts_required
    false

    endpoints

    endpoints: 3
    • GET /v1/models/baobab-yo
    • GET /v1/baobab/yo/national
    • POST /v1/demo/speech

    limits

    translate_chars_daily_hint
    Authenticated translate required for production MT
    demo_speech
    400 requests / 10 minutes / IP (demo controller)
    rate_limit_note
    Demo speech returns rate_limited when exceeded

    controls

    controls: 3
    • honest en-NG remap when speech offline (paired Nigerian English accent)
    • spectral_gate_v2 clarity on live neural WAV
    • variety headers — yo-ng

  • Education / free

    id
    education
    name
    Education / free
    who
    Students, UG / KNUST labs, research orgs on Free/Pro plans
    auth
    Org API key (TranslateAuthGuard) — Free or Pro plan
    maps_to_plan
    free|pro
    residency_preference
    NG
    audit
    true
    commercial_nc_clean_tts_required
    false

    endpoints

    endpoints: 4
    • POST /v1/translate
    • POST /v1/tts/synthesize
    • POST /v1/audio/transcriptions (when ASR keyed)
    • GET /v1/models/baobab-yo

    limits

    translate_chars_daily_hint
    Free: BILLING_FREE_TRANSLATE_CHARS_QUOTA (default 50k); Pro: BILLING_PRO_TRANSLATE_CHARS_QUOTA
    demo_speech
    null
    rate_limit_note
    RATE_LIMIT_FREE_PER_KEY / RATE_LIMIT_PRO_PER_KEY via existing RateLimitGuard

    controls

    controls: 2
    • UsageService metering
    • BillingService.assertWithinQuota

  • Ministry production

    id
    ministry
    name
    Ministry production
    who
    Nigeria government services (Lagos / Southwest Nigeria citizen channels)
    auth
    Ministry-keyed org API key — Business plan mapping
    maps_to_plan
    business
    residency_preference
    NG
    audit
    true
    commercial_nc_clean_tts_required
    false

    endpoints

    endpoints: 5
    • POST /v1/translate
    • POST /v1/tts/synthesize
    • MeetingBridge Ghana corridor
    • GET /v1/residency
    • GET /v1/metrics/translate

    limits

    translate_chars_daily_hint
    Business translate quota (BILLING_BUSINESS_TRANSLATE_CHARS_QUOTA)
    demo_speech
    null
    rate_limit_note
    RATE_LIMIT_BUSINESS_PER_KEY / PER_ORG

    controls

    controls: 3
    • AuditService on translate.completed
    • Org retention / vendor-training settings
    • Residency preference GH (affinity, not hard geo-fence)

  • Enterprise

    id
    enterprise
    name
    Enterprise
    who
    Banks, telcos, NGOs serving Lagos / Southwest Nigeria
    auth
    Paid Enterprise keys
    maps_to_plan
    enterprise
    residency_preference
    AF
    audit
    true
    commercial_nc_clean_tts_required
    true

    endpoints

    endpoints: 3
    • POST /v1/translate (+ TM / quality)
    • Speech + MeetingBridge
    • Fine-tunes / marketplace where plan allows

    limits

    translate_chars_daily_hint
    Enterprise translate quota
    demo_speech
    null
    rate_limit_note
    RATE_LIMIT_ENTERPRISE_PER_KEY / PER_ORG

    controls

    controls: 3
    • SLA targets (documented; contractual SLA separate)
    • Commercial TTS path must be NC-clean for redistribution of MMS TTS assets
    • Audit + retention

research

id
baobab-yo-generative-lm
status
research_planned
foundation_llm_weights
false
generative_inference_live
false
ships_trained_baobab_foundation
false
separate_variety_checkpoints_required
true
training_recipe
docs/baobab/GENERATIVE_LM_RECIPE.md
note
Research slot + harness + training recipe only. Do not brand phrase MT or speech corridors as a full national LLM. Generative inference unsupported until gates are met.

proposed_checkpoint_ids

primary
baobab-yo-lagos-lm-*

prerequisites

prerequisites: 3
  • en-yo-lagos corridor + national API surfaces remain the product wedge
  • Mix / Fidelity / Live evals and rights clear
  • Consented parallel + speech data with held-outs

acceptance_gates

acceptance_gates: 5

  • held_out_speakers

    id
    held_out_speakers
    required
    true
    met
    false
    description
    Held-out evaluation speakers must never appear in training exports; speaker lists remain empty until consented collection.

  • community_consent

    id
    community_consent
    required
    true
    met
    false
    description
    trainingEligible stays false until Lagos / Southwest Nigeria community speakers re-confirm for training.

  • variety_separation

    id
    variety_separation
    required
    true
    met
    false
    description
    Any future LM/adapters must ship as baobab-yo-lagos-* — never one blob marketed as interchangeable varieties.

  • beat_phrase_mt_baselines

    id
    beat_phrase_mt_baselines
    required
    true
    met
    false
    description
    Generative claims require held-out gains vs corridor phrase-MT baselines on declared native tasks.

  • commercial_rights_clear

    id
    commercial_rights_clear
    required
    true
    met
    false
    description
    Base model + data licenses must clear commercial ministry/enterprise redistribution before promotion.

task_harness

goldens
apps/api/src/baobab/baobab-yo.research.ts#RESEARCH_GOLDENS
runner
apps/api/src/baobab/baobab-yo.research.ts#evaluateBaobabYoNativeTasks
generative_default
unsupported

training_eligible_unlock

trainingEligible
false
status
locked
unlock_doc
docs/baobab/TRAINING_ELIGIBLE_UNLOCK.md
note
trainingEligible remains false until all gates are met with durable evidence. Do not invent consent.

gates

gates: 5

  • community_consent_yo

    id
    community_consent_yo
    required
    true
    met
    false
    evidence_required
    Signed Lagos / Southwest Nigeria contributor agreements reconfirming training use
    description
    Lagos / Southwest Nigeria community speakers must reconfirm training export consent.

  • held_out_speaker_lists

    id
    held_out_speaker_lists
    required
    true
    met
    false
    evidence_required
    Published held-out speaker IDs never present in train splits
    description
    Held-out evaluation speakers must never appear in training exports.

  • yo_native_review

    id
    yo_native_review
    required
    true
    met
    false
    evidence_required
    Paid Yoruba (Lagos / Oyo) native review records with trainingEligible fields
    description
    Native review must cover Yoruba (Lagos / Oyo) before any train export.

  • variety_not_conflated

    id
    variety_not_conflated
    required
    true
    met
    false
    evidence_required
    Corpus metadata proving non-interchangeable varieties excluded from train exports
    description
    Egba (yo-ng-egba) is a secondary Nigerian variety — not the Ghana national default.

  • rights_clear_for_export

    id
    rights_clear_for_export
    required
    true
    met
    false
    evidence_required
    Legal memo clearing commercial vs research split for export corpus
    description
    IP / commercial rights must clear before trainingEligible flips.

license_warnings

license_warnings: 4
  • Yoruba MMS-TTS (facebook/mms-tts-yor) is CC-BY-NC-4.0 — unrestricted commercial redistribution of those assets is blocked.
  • Whisper ASR for Yoruba corridor — pin versions and attribute per model card.
  • Lagos / Oyo is the product default via prosody + labels on shared Yoruba-general weights — not a Ghana-only exclusive speech finetune.
  • Phrase/lexical Baobab MT covers reviewed corridor domains; neural_nmt is false (no Hausa NLLB national service shipped).

capabilities

phrase_mt
phrase_table_only
national_api_tiers
true
generative_lm_research_slot
true
generative_inference_live
false
neural_nmt
false
neural_asr
false
neural_tts
false
s2s_corridor
true
foundation_llm_weights
false
audio_clean
spectral_gate_v2

honesty

phrase_table_only
true
national_api_tiers
true
research_slot_shipped
true
neural_speech_live
false
neural_nmt
false
generative_inference_live
false
foundation_llm_weights
false
ships_trained_baobab_foundation
false
variety_exclusive_speech_finetunes
false
trainingEligible
false
iso_code
yor
audio_clean
spectral_gate_v2

training_eligible_unlock

trainingEligible
false
status
locked
unlock_doc
docs/baobab/TRAINING_ELIGIBLE_UNLOCK.md
note
trainingEligible remains false until all gates are met with durable evidence. Do not invent consent.

gates

gates: 5

  • community_consent_yo

    id
    community_consent_yo
    required
    true
    met
    false
    evidence_required
    Signed Lagos / Southwest Nigeria contributor agreements reconfirming training use
    description
    Lagos / Southwest Nigeria community speakers must reconfirm training export consent.

  • held_out_speaker_lists

    id
    held_out_speaker_lists
    required
    true
    met
    false
    evidence_required
    Published held-out speaker IDs never present in train splits
    description
    Held-out evaluation speakers must never appear in training exports.

  • yo_native_review

    id
    yo_native_review
    required
    true
    met
    false
    evidence_required
    Paid Yoruba (Lagos / Oyo) native review records with trainingEligible fields
    description
    Native review must cover Yoruba (Lagos / Oyo) before any train export.

  • variety_not_conflated

    id
    variety_not_conflated
    required
    true
    met
    false
    evidence_required
    Corpus metadata proving non-interchangeable varieties excluded from train exports
    description
    Egba (yo-ng-egba) is a secondary Nigerian variety — not the Ghana national default.

  • rights_clear_for_export

    id
    rights_clear_for_export
    required
    true
    met
    false
    evidence_required
    Legal memo clearing commercial vs research split for export corpus
    description
    IP / commercial rights must clear before trainingEligible flips.

nmt

neural_nmt
false
disclosed_base_named
null
disclosed_base_weights_in_repo
false
commercial_license_cleared
false
adapter_status
phrase_table_only
checkpoint_shipped
false
phrase_table_live
true
remote_mt_configured
false
remote_mt_env
null
baobab_yo_nmt_url_configured
false
baobab_yo_nmt_env
null
separate_variety_adapters
false
note
Baobab Yoruba MT is phrase/lexical only (en↔yo). Fail-closed when uncovered. neural_nmt false.

phrase_table_pairs

phrase_table_pairs: 1
  • en↔yo

residency

serving_today
jnb
serving_today_label
Africa-primary Fly region Johannesburg (jnb)
affinity_target
af-west-1
affinity_city
Lagos
affinity_country
NG
accra_gpu_island
false
accra_gpu_island_status
deferred
hard_geo_fence
false
preference_model
affinity_not_hard_fence
ministry_affinity_pin
af-west-1
affinity_enforcement
soft_preference_recorded
accra_island_runbook
docs/baobab/ACCRA_RESIDENCY.md
note
Inference prefers matching person/org residency to nearby model hosting when ready — affinity, not a hard geo-fence. Lagos is the Nigeria residency preference; Africa-west affinity pin remains af-west-1. Live serving for verbalab + lugemi-yoruba is jnb today. No Accra GPU island is deployed.

licenses

tts
CC-BY-NC-4.0
asr
see_model_card
note
TTS is CC-BY-NC-4.0. neural_nmt stays false — no dedicated NLLB remote shipped. Clarity path: spectral_gate_v2.

deferred

deferred: 6
  • Commercially cleared neural NMT (phrase MT is the live fail-closed path)
  • Accra GPU island / hard Ghana geo-fence (affinity af-west-1; serving jnb)
  • Contractual ministry SLA signature (operational targets shipped)
  • Generative LM weights + inference (recipe + gates only)
  • Variety-exclusive ASR/TTS finetunes beyond shared checkpoints
  • trainingEligible unlock evidence (checklist shipped; flag locked false)

endpoints

model_card
GET /v1/models/baobab-yo
national
GET /v1/baobab/yo/national
research
GET /v1/baobab/yo/research
unlock
GET /v1/baobab/yo/unlock
slo
GET /v1/baobab/yo/slo
translate
POST /v1/translate
demo_speech
POST /v1/demo/speech
tts
POST /v1/tts/synthesize
meetingbridge_ghana
MeetingBridge Ghana corridor
corridor
en-yo-lagos
residency
GET /v1/residency
translate_metrics
GET /v1/metrics/translate

docs

national_model
docs/baobab/YORUBA_NATIONAL_MODEL.md
model_card
docs/baobab/baobab-yo.MODEL_CARD.md
partnership_open_eval
docs/baobab/PARTNERSHIP_OPEN_EVAL.md
ministry_sla
docs/baobab/MINISTRY_SLA.md
training_eligible_unlock
docs/baobab/TRAINING_ELIGIBLE_UNLOCK.md
speech_finetune_recipe
docs/baobab/SPEECH_FINETUNE_RECIPE.md
generative_lm_recipe
docs/baobab/GENERATIVE_LM_RECIPE.md
accra_residency
docs/baobab/ACCRA_RESIDENCY.md
corridor
en-yo-lagos
language_kit
docs/models/05_LANGUAGE_KIT.md
residency
docs/residency.md