Google Speech Recognition

OpenMind proxies Google Speech Recognition (ASR) API to provide speech recognition capabilities. This endpoint allows you to interact with the Google ASR API to transcribe speech to text.

To minimize latency, the API endpoint utilizes WebSockets for efficient real-time communication.

wss://api.openmind.org/api/core/google/asr?api_key=<YOUR_API_KEY>

​

Installation

Install the OM1 package:

uv pip3 install git+https://github.com/OpenmindAGI/OM1.git

If you don’t have uv installed, you can install it using the following command:

pip3 install git+https://github.com/OpenmindAGI/OM1.git

​

Usage

The following example demonstrates how to interact with the Google ASR API using the OM1 package:

import time
from om1_utils import ws
from om1_speech import AudioInputStream

# Initialize the Google ASR API
ws_client = ws.Client(url="wss://api.openmind.org/api/core/google/asr?api_key=<YOUR_API_KEY>")
audio_stream_input = AudioInputStream(audio_data_callback=ws_client.send_message)

# Start the Google ASR API
ws_client.start()
audio_stream_input.start()

# Retrieve the Google ASR API response
ws_client.register_message_callback(lambda msg: print(msg))

while True:
  time.sleep(1)

The expected response from the Google ASR API will be in the following format:

{
  "asr_reply": "hello world"
}

You can also forward the base64 encoded audio data directly to the API endpoint using the following format:

{
  "audio": "base64_encoded_audio_data",
  "rate": 16000
}