Sign InSign Up

Best-in-class Speech-to-Text API

Speech to Text Performance Comparison

This chart demonstrates how our Speech-to-Text API outperforms competitors in terms of accuracy and processing speed. Choose the model that best fits your use case.

Available Models

  • • en-US-0.6b
  • • em-ea-1.1b
  • • en-US-NER
  • • en-US-300m
  • • hi-IN-NER
  • • ia-IN-NER
  • • ru-RU-300m
  • • en-US-IoT-NER

Code Examples


import requests

# Function to list available ASR models
def list_asr_models(auth_token):
    url = 'https://api.whissle.ai/v1/list-asr-models'
    headers = {'Authorization': f'Bearer {auth_token}'}
    response = requests.get(url, headers=headers)
    return response.json() if response.status_code == 200 else response.text

# Function to transcribe audio using a specific ASR model
def transcribe_audio(auth_token, model_name, audio_path):
    url = f'https://api.whissle.ai/v1/conversation/STT?model_name={model_name}&auth_token={auth_token}'
    files = {'audio': open(audio_path, 'rb')}
    response = requests.post(url, files=files)
    return response.json() if response.status_code == 200 else response.text

# Example usage
if __name__ == "__main__":
    AUTH_TOKEN = "your_auth_token_here"  # Replace with your token
    print("Available Models:", list_asr_models(AUTH_TOKEN))
    transcription = transcribe_audio(AUTH_TOKEN, "en-US-0.6b", "path/to/audio.wav")
    print("Transcription:", transcription)