ASR & Transcription Functional Overview & API Integration
Capacity Private Cloud ASR converts speech to text — both grammar-based recognition, which matches an utterance against a defined grammar and returns a structured semantic result, and free-form transcription, which returns the spoken words verbatim. This page is a starting point: it routes you to the product overview, the supporting reference articles, and the developer API definitions.
Start here — ASR & Transcription overview
Speech Products: ASR & Transcription — what the recognition and transcription products do, where each fits, and the capabilities they expose.
Functional Overview & Background
These reference articles cover the grammar and semantic-interpretation concepts behind recognition. They are the foundation for writing grammars that return the results your application expects.
| Article | What it covers |
| An SRGS Introduction | The Speech Recognition Grammar Specification — the standard used to define what the recognizer listens for. |
| A Simple Grammar | A worked first grammar, from structure to a working recognition result. |
| Using GrXML Grammars | Working with the XML form of SRGS grammars. |
| Intro to Semantic Interpretation | How semantic interpretation (SI) tags turn a recognized phrase into a structured result. |
| SISR Basics | The Semantic Interpretation for Speech Recognition standard and how to use it in grammars. |
Developer API Reference
Use the following information when implementing directly against the gRPC API, configuring recognition, or wiring up reporting.
| Reference | Description |
| ASR integration | API integration guide for ASR and transcription. |
asr-lumenvox.proto | gRPC API definition for ASR and transcription interactions. |
| ASR configuration | Configuration settings for recognition and transcription. |
reporting.proto | gRPC API definition for usage and results reporting. |
