POST https://api.meta.ai/v1/asr/transcribe with Bearer header and multipart request JSON plus audio WAV. Optional sessionId query value is a correlation identifier, not authentication.
Regions
Availability is geography-controlled; see Meta geographic-use policy. No data-residency guarantee is inferred.
SSE has no appended final JSON summary. Plain text can include speaker-prefixed lines; JSON has empty turns for PUSH_TO_TALK.
Prices retain their published currencies and billing units below.
Published billing units · original currency · observed 13 Oct 2026
Category
Usage
Rate
Unit
Speech recognition
Audio processed1
$0.18
hour
Rate conditions
Audio actually processed, rounded down to whole seconds. File and realtime cost the same; ZDR at parity. Pre-transcript failures and rate-limited requests are not billed according to the speech guide.
Standard requests are not used to train Meta models. Terms allow retention for service operation, safety/security, legal and policy purposes without a fixed general duration. Standard mode is not itself a zero-retention guarantee.
wss://api.meta.ai/v1/asr/realtime. First JSON handshake within 10 seconds supplies authorization.accessToken="Bearer <key>". HTTP Authorization headers and API-key query parameters are not accepted for authentication.
Regions
Availability is geography-controlled; see Meta geographic-use policy. No data-residency guarantee is inferred.
Send endStream after draining queued audio, then keep reading until server closes with 1000. No per-turn commit or resume token.Source for completion (opens in a new tab)
Prices retain their published currencies and billing units below.
Published billing units · original currency · observed 13 Oct 2026
Category
Usage
Rate
Unit
Speech recognition
Audio processed1
$0.18
hour
Rate conditions
Audio actually processed, rounded down to whole seconds. File and realtime cost the same; ZDR at parity. Pre-transcript failures and rate-limited requests are not billed according to the speech guide.
Standard requests are not used to train Meta models. Terms allow retention for service operation, safety/security, legal and policy purposes without a fixed general duration. Standard mode is not itself a zero-retention guarantee.
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.
Meta Model API
muse-voice-transcribe-1.0
Not verified
No endpoint-specific discounted prompt-cache contract verified.
How it works
Not collected
No endpoint-specific discounted prompt-cache contract verified.
Minimum reusable input
Not collected
No endpoint-specific discounted prompt-cache contract verified.
Lifetime & refresh
Not collected
No endpoint-specific discounted prompt-cache contract verified.
How to enable it
Not collected
No endpoint-specific discounted prompt-cache contract verified.
Usage fields
Not collected
No endpoint-specific discounted prompt-cache contract verified.
Billing rules
Not collected
No endpoint-specific discounted prompt-cache contract verified.
Limits & data boundaries
Not collected
No endpoint-specific discounted prompt-cache contract verified.
No verified cache rate for this endpoint. An empty rate is not a zero charge.
Meta Model API
muse-voice-transcribe-1.0
Not verified
No endpoint-specific discounted prompt-cache contract verified.
How it works
Not collected
No endpoint-specific discounted prompt-cache contract verified.
Minimum reusable input
Not collected
No endpoint-specific discounted prompt-cache contract verified.
Lifetime & refresh
Not collected
No endpoint-specific discounted prompt-cache contract verified.
How to enable it
Not collected
No endpoint-specific discounted prompt-cache contract verified.
Usage fields
Not collected
No endpoint-specific discounted prompt-cache contract verified.
Billing rules
Not collected
No endpoint-specific discounted prompt-cache contract verified.
Limits & data boundaries
Not collected
No endpoint-specific discounted prompt-cache contract verified.
No verified cache rate for this endpoint. An empty rate is not a zero charge.
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
Meta identifies this native Muse model in its research or hosted-model catalog. The reviewed release is available through provider-controlled service, subscription or research preview, without a released weight artifact in the reviewed inventory. The September Spark roadmap still describes its weight release as forthcoming. Open-weight Glimmer, SAM and Llama are excluded separately; hosted access alone does not turn those models into closed models.