1.1.0 • Published 8 months ago

@aristech-org/stt-client v1.1.0

Weekly downloads
-
License
MIT
Repository
github
Last release
8 months ago

Aristech STT-Client for NodeJS

This is the NodeJS client implementation for the Aristech STT-Server.

Installation

npm install @aristech-org/stt-client

Usage

import { SttClient } from '@aristech-org/stt-client'

const client = new SttClient({ host: 'stt.example.com' })
const results = await client.recognizeFile('path/to/audio/file.wav')
console.log(results.map(r => r.chunks[0].alternatives[0].text).join('\n'))

There are several examples in the examples directory:

  • recognize.ts: Pretty much the same as the example above.
  • streaming.ts: Demonstrates how to stream audio to the server while receiving interim results.
  • models.ts: Demonstrates how to get the available models from the server.
  • nlpFunctions.ts: Demonstrates how to list the configured NLP-Servers and the coresponding functions.
  • nlpProcess.ts: Demonstrates how to perform NLP processing on a text by using the STT-Server as a proxy.
  • account.ts: Demonstrates how to retrieve the account information from the server.

You can run the examples directly using tsx like this:

  1. Create a .env file in the node directory:
HOST=stt.example.com
# The credentials are optional but probably required for most servers:
TOKEN=your-token
SECRET=your-secret

# The following are optional:
# ROOT_CERT=your-root-cert.pem # If the server uses a self-signed certificate
# SSL=true # Set to true if credentials are provided or if a ROOT_CERT is provided
# MODEL=some-available-model
# NLP_SERVER=some-config
# NLP_PIPELINE=function1,function2
  1. Run the examples, e.g.:
npx tsx examples/streaming.ts

Build

To rebuild the generated typescript files from the proto file, run:

npm run generate

To build the library, run:

npm run build
1.1.0

8 months ago

1.0.3

9 months ago

1.0.2

9 months ago

1.0.1

9 months ago

1.0.0

9 months ago