No reviewed model-specific value; unknown does not mean unsupported.
Fine-tuning
Not collected
No reviewed model-specific value; unknown does not mean unsupported.
Reasoning
Not collected
No reviewed model-specific value; unknown does not mean unsupported.
Context caching
Not collected
No reviewed model-specific value; unknown does not mean unsupported.
Base model
Not collected
Public 9jaLingo-TTS-African-ckpt-25k is Spark-TTS/SparkVox-based with Qwen2.5-0.5B-Instruct, BiCodec and wav2vec2-large-xlsr-53. Mapping to hosted 9jalingo-tts-1 is not proven; do not inherit architecture or weights.
SDK sends enable_long_form=true by default with optional max_chunk_duration and silence_duration; server bounds for those options are not published. Omitted voice defaults to blessing_pcm unless a speaker embedding is provided.Source for Interactive controls (opens in a new tab)
Account-scoped reusable custom voices, not separate model releases.
Reference audio
At least 30 seconds of clear reference audio. Documented clone formats include WAV, MP3, M4A, FLAC and OGG; a complete clone-specific format/size ceiling is not established.Source for Reference audio (opens in a new tab)
Custom voice retention
Clone returns X-Voice-ID, X-Voice-Code and X-Clone-ID; use the voice ID for later synthesis/streaming. DELETE /v1/voices/{voice_id} removes an owned clone with HTTP 204. No verified purge TTL.Source for Custom voice retention (opens in a new tab)
Request format
Synthesis: JSON input/voice/lang/model. Clone: multipart text/audio with language and optional speaker. Public cURL uses voice as a language code; the pinned SDK normalizes language into lang and uses voice only for a resolved speaker.Source for Request format (opens in a new tab)
Published price
One-off credit bundles: NGN 5,000 buys 10,000 credits; NGN 10,000 buys 20,000; NGN 50,000 buys 110,000 including 10,000 bonus. Starter Gift grants 500 credits. Published NGN tariff equivalents use the standard schedule; bundle bonuses change effective credit acquisition cost. Do not assume a universal credit-to-currency conversion.Source for Published price (opens in a new tab)
Usage metering
Standard: 100 credits or NGN 50 per 1,000 characters. Custom-voice generation: 102 credits or NGN 51 per 1,000 characters. New voice-profile setup is additional.Source for Usage metering (opens in a new tab)
Starter card says 30-day credit rollover; the same page says credits never expire while the account is active. FAQ also refers to monthly allowances despite one-off bundles. Exact expiry/allowance contract needs confirmation.Source for Documentation discrepancy (opens in a new tab)
Access & pricing
Prices belong to an endpoint and billing tier. A provider listing does not guarantee access for every account or region.
Service selector; immutable backing checkpoint not established.
Access
POST https://api.9jalingo.org/v1/audio/speech; JSON input and voice/speaker options, X-API-Key. Public metadata inventory lists 9jalingo-tts-1.
Regions
Inference hosting region not established; company location does not prove processing residency.
Prices retain their published currencies and billing units below.
Published billing units · original currency · observed 16 Oct 2026
Category
Usage
Rate
Unit
Standard TTS
Built-in voice generation1
100 credits
1,000 characters
Custom-voice generation
Custom voice generation (includes 2% premium)2
102 credits
1,000 characters
Standard TTS
Built-in voice generation1
NGN 50.00
1,000 characters
Custom-voice generation
Custom voice generation (includes 2% premium)2
NGN 51.00
1,000 characters
Rate conditions
Prepaid PAYG credits. NGN amounts are the provider’s published standard tariff equivalents; bonus bundles change effective acquisition cost. Text character counting, minimum charge and billing rounding are not specified. No text-token conversion.
Prepaid PAYG credits. NGN amounts are the provider’s published standard tariff equivalents; bonus bundles change effective acquisition cost. Text character counting, minimum charge and billing rounding are not specified. No text-token conversion. Reusing an existing clone incurs the custom-voice generation rate; extraction/storage setup fee is separate.
Privacy permits processing submitted text/audio to provide services and using information to improve AI systems; it does not establish a blanket no-training guarantee. Retention is purpose-based without a fixed TTL, and users can request deletion. Hosting/payment/analytics providers may receive necessary data. The pricing FAQ says generated audio and voice profiles can be deleted, but its broad no-sharing language does not remove the privacy policy vendor disclosures. No exact inference residency, zero-retention contract or independent security certification established.
No model selector sent on this route in the reviewed contract.
Access
POST https://api.9jalingo.org/v1/audio/speech/stream; X-API-Key. SDK sends WAV output and current default service, with no model field. Do not treat the default as an immutable pin.
Regions
Inference hosting region not established; company location does not prove processing residency.
Prices retain their published currencies and billing units below.
Published billing units · original currency · observed 16 Oct 2026
Category
Usage
Rate
Unit
Standard TTS
Built-in voice generation1
100 credits
1,000 characters
Custom-voice generation
Custom voice generation (includes 2% premium)2
102 credits
1,000 characters
Standard TTS
Built-in voice generation1
NGN 50.00
1,000 characters
Custom-voice generation
Custom voice generation (includes 2% premium)2
NGN 51.00
1,000 characters
Rate conditions
Prepaid PAYG credits. NGN amounts are the provider’s published standard tariff equivalents; bonus bundles change effective acquisition cost. Text character counting, minimum charge and billing rounding are not specified. No text-token conversion.
Prepaid PAYG credits. NGN amounts are the provider’s published standard tariff equivalents; bonus bundles change effective acquisition cost. Text character counting, minimum charge and billing rounding are not specified. No text-token conversion. Reusing an existing clone incurs the custom-voice generation rate; extraction/storage setup fee is separate.
Privacy permits processing submitted text/audio to provide services and using information to improve AI systems; it does not establish a blanket no-training guarantee. Retention is purpose-based without a fixed TTL, and users can request deletion. Hosting/payment/analytics providers may receive necessary data. The pricing FAQ says generated audio and voice profiles can be deleted, but its broad no-sharing language does not remove the privacy policy vendor disclosures. No exact inference residency, zero-retention contract or independent security certification established.
Service selector; immutable backing checkpoint not established.
Access
POST https://api.9jalingo.org/v1/audio/clone; multipart reference audio/text, X-API-Key, SDK model_name=9jalingo-tts-1. Returned owned voice ID can be reused on synthesis or stream routes; deletion is a separate management action.
Regions
Inference hosting region not established; company location does not prove processing residency.
Prices retain their published currencies and billing units below.
Published billing units · original currency · observed 16 Oct 2026
Category
Usage
Rate
Unit
Custom-voice generation
Custom voice generation (includes 2% premium)1
102 credits
1,000 characters
Custom-voice generation
Custom voice generation (includes 2% premium)1
NGN 51.00
1,000 characters
Voice profile add-on
One-time voice embedding extraction and storage2
50 credits
new voice profile
Rate conditions
Prepaid PAYG credits. NGN amounts are the provider’s published standard tariff equivalents; bonus bundles change effective acquisition cost. Text character counting, minimum charge and billing rounding are not specified. No text-token conversion. Voice-profile setup is additional.
Pricing FAQ states a one-time 50-credit fee for extracting and storing a reusable voice profile. Generated speech is charged separately. No separately published NGN setup tariff assigned.
Privacy permits processing submitted text/audio to provide services and using information to improve AI systems; it does not establish a blanket no-training guarantee. Retention is purpose-based without a fixed TTL, and users can request deletion. Hosting/payment/analytics providers may receive necessary data. The pricing FAQ says generated audio and voice profiles can be deleted, but its broad no-sharing language does not remove the privacy policy vendor disclosures. No exact inference residency, zero-retention contract or independent security certification established.
No benchmark observations have been added for this model. We do not turn a vendor description into a performance score.
Latency and throughput depend on the host, request size, reasoning settings and load. There is no measured speed or global usage figure in this record.
The record behind the model
Why it is included
Commercial hosted speech service under proprietary platform terms. This is not a claim of private foundation weights: the provider also publishes Apache-2.0 Spark-TTS-derived artifacts, and exact hosted checkpoint equivalence is unverified.
