Google Cloud Speech-to-text API

An example of server an API with Express, using the Google Cloud Speech-to-text API to transcribe audio.

You can test the API with Postman, and call the API with form-data body and the following params.

API Reference

POST /api/sts

Request Params

• audio: An audio file with wav/linear format, and the sample rate is 44100.

Response

• transcription: Text transcription
• confidence: Confidence score

Getting Started

First, to install the dependices run:

yarn install

Then, export the credentials for the Google Cloud Platform on CLI:

export GOOGLE_APPLICATION_CREDENTIALS=<path to the json file>

Then, build and run the server:

yarn build
yarn start

Configuration

Create a file called .env on the root directory and add the following to change the port and sample rate.

PORT=<port>
SAMPLE_RATE=<sample rate>