twi-en: Akan Twi + English (Recommended). Best for Twi, English, and Twi-English code-switching., twi-only: Akan Twi (No English). Use only when the entire recording is spoken in Twi. English words may be missed., en: English (Ghanaian accent). English speech with a Ghanaian accent., gpe: Ghanaian Pidgin English. Ghanaian Pidgin English speech.Source for Languages (opens in a new tab)
Output formats
JSON text, transcription, language, duration_seconds, credits_remaining and credit_limit. Optional reference_text enables per-request WER/accuracy.Source for Output formats (opens in a new tab)
Published price
Terms list Pro $25/month with 5,000 credits and Enterprise $199/month with 50,000 credits. Credits do not roll over. ISO currency and task-level credit conversion are not established; these are subscription totals, not per-token or per-question prices.Source for Published price (opens in a new tab)
Request quotas
25 free transcriptions without sign-in; 50 total after sign-in, not 75. Quota reset window and paid task-credit conversion are unspecified.Source for Request quotas (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.
voice or language selects a task option, not a documented immutable checkpoint.
Access
POST https://abena.mobobi.com/playground/api/v1/asr/transcribe/ (multipart audio_file, optional language and reference_text). Free usage requires no API key; authenticated use accepts Authorization: Bearer or X-API-Key.
Regions
Hosting location not disclosed; policy allows international processing.
Endpoint pricing has not been collected. Check the provider for applicable rates.
Data & enterprise terms
Hosted API privacy says audio/text are deleted after processing and are not used for training without explicit consent. Account, billing, request, performance and error logs are separately collected. Exact log retention and hosting region are unspecified; international processing and trusted service providers are allowed. This does not establish universal zero retention.
Caching can reuse earlier input processing. Availability, lifetime and billing depend on the endpoint. A cache hit still generates a new response.
01 / INITIAL INPUTProcess & retain
The system processes eligible input and retains reusable state.
02 / LATER REQUESTCheck available state
Reuse depends on matching input, session state and the cache policy.
03 / CACHE HITReuse & generate
Reusing state can reduce repeated work. Any charges follow the billing rules below.
Abena hosted API
Not verified
No native inference-cache contract reviewed. Server model loading and offline operation are not evidence of prompt caching.
How it works
Not collected
No native inference-cache contract reviewed. Server model loading and offline operation are not evidence of prompt caching.
Minimum reusable input
Not collected
No native inference-cache contract reviewed. Server model loading and offline operation are not evidence of prompt caching.
Lifetime & refresh
Not collected
No native inference-cache contract reviewed. Server model loading and offline operation are not evidence of prompt caching.
How to enable it
Not collected
No native inference-cache contract reviewed. Server model loading and offline operation are not evidence of prompt caching.
Usage fields
Not collected
No native inference-cache contract reviewed. Server model loading and offline operation are not evidence of prompt caching.
Billing rules
Not collected
No native inference-cache contract reviewed. Server model loading and offline operation are not evidence of prompt caching.
Limits & data boundaries
Not collected
No native inference-cache contract reviewed. Server model loading and offline operation are not evidence of prompt caching.
No verified cache rate for this endpoint. An empty rate is not a zero charge.
Caching reviewed 16 Oct 2026
What the measurements say
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 speech capability with undisclosed backing checkpoints. API terms reserve model ownership and prohibit extraction; the linked MIT repository supplies client examples, not model weights. This service entry does not establish that every backing artifact is independently closed-weight.
