Transcription service(s) management

Our LinTO Platform Service Manager is a service that manages the scalabitily of linTO Platform STT services depending on the number of queries (scale up or down the number of replicas of each service). Docker swarm is used as a container orchestration tool, meaning that it allows the user to manage multiple containers deployed across multiple host machines. Along with that, this service is designed to manage the linTO Platform STT decoding models, i.e. acoustic models and the decoding graph.

Summary

• How2use: APIs definition and use guidelines

📄️ Create an acoustic model

Acoustic model

📄️ Deploy a command transcription service

Command transcription service

📄️ Deploy a Large vocabulary file transcription service

A file transcription service allows to submit audio files to an API that will return a text transcription. File transcription can be used as itself and is a prerequisite of other uses. "Large vocabulary" implies that the transcription won't be limited to preset sentences and intend to cover an entire language.

📄️ Deploy a Large Vocabulary Streaming Transcription service

A streaming transcription service deploys a API that accept an audio stream and return a real-time transcription. "Large vocabulary" implies that the transcription won't be limited to preset sentences and intend to cover an entire language..

📄️ Streaming Transcription service usage

Streaming speech to text standalone worker. Processes audio streams on an WS endpoint

📄️ Platform Service manager

Platform Service manager is a service designed to create, update and scale Speech-To-Text service inside your linTo platform. In this documentation you will be provided with: