85 lines
2.5 KiB
Python
85 lines
2.5 KiB
Python
|
from typing import Optional, Union, Dict, Any
|
||
|
|
|
||
|
|
import requests
|
||
|
|
|
||
|
|
|
||
|
|
def send_asr_request(url: str, audio_file_path_or_bytes: str | bytes, task: Optional[str] = None, language: Optional[str] = None,
|
||
|
|
initial_prompt: Optional[str] = None, encode: Optional[bool] = None,
|
||
|
|
output: Optional[str] = None, word_timestamps: Optional[bool] = None) -> tuple[int, str]:
|
||
|
|
"""
|
||
|
|
Send a request to the ASR endpoint.
|
||
|
|
Returns the text represented by the audio file if everything worked out,
|
||
|
|
and a tuple of the form (status_code, response_text) otherwise
|
||
|
|
"""
|
||
|
|
endpoint = f"{url}/asr"
|
||
|
|
|
||
|
|
params = {
|
||
|
|
"task": task,
|
||
|
|
"language": language,
|
||
|
|
"initial_prompt": initial_prompt,
|
||
|
|
"encode": encode,
|
||
|
|
"output": output,
|
||
|
|
"word_timestamps": word_timestamps
|
||
|
|
}
|
||
|
|
|
||
|
|
params = {k: v for k, v in params.items() if v is not None}
|
||
|
|
|
||
|
|
if isinstance(audio_file_path_or_bytes, str):
|
||
|
|
with open(audio_file_path_or_bytes, 'rb') as f:
|
||
|
|
audio_file = f.read()
|
||
|
|
else:
|
||
|
|
audio_file = audio_file_path_or_bytes
|
||
|
|
|
||
|
|
files = {
|
||
|
|
'audio_file': audio_file
|
||
|
|
}
|
||
|
|
|
||
|
|
response = requests.post(endpoint, params=params, files=files)
|
||
|
|
|
||
|
|
return response.status_code, response.text
|
||
|
|
|
||
|
|
|
||
|
|
def detect_language(url: str, audio_file_path: str, encode: Optional[bool] = None) -> Dict[str, Any] | tuple[int, str]:
|
||
|
|
"""
|
||
|
|
Send a request to the Detect Language endpoint.
|
||
|
|
Returns either a dictionary of the form {'detected_language': '<LANG>', 'language_code': '<LANG_CODE>'} if the request
|
||
|
|
was successful, or a tuple of the form (status_code, response_text) otherwise.
|
||
|
|
"""
|
||
|
|
endpoint = f"{url}/detect-language"
|
||
|
|
|
||
|
|
params = {
|
||
|
|
"encode": encode
|
||
|
|
}
|
||
|
|
|
||
|
|
params = {k: v for k, v in params.items() if v is not None}
|
||
|
|
|
||
|
|
with open(audio_file_path, 'rb') as f:
|
||
|
|
audio_file = f.read()
|
||
|
|
|
||
|
|
files = {
|
||
|
|
'audio_file': audio_file
|
||
|
|
}
|
||
|
|
|
||
|
|
response = requests.post(endpoint, params=params, files=files)
|
||
|
|
|
||
|
|
if response.status_code == 200:
|
||
|
|
return response.json()
|
||
|
|
else:
|
||
|
|
return response.status_code, response.text
|
||
|
|
|
||
|
|
|
||
|
|
# Example usage
|
||
|
|
def main():
|
||
|
|
url = "http://127.0.0.1:9000" # Replace with the actual URL of the webservice
|
||
|
|
audio_file_path = "/run/media/yannik/IC RECORDER/REC_FILE/Interview01/231021_1541.mp3"
|
||
|
|
|
||
|
|
response = send_asr_request(url, audio_file_path, task="transcribe", language="en")
|
||
|
|
print(response)
|
||
|
|
|
||
|
|
response = detect_language(url, audio_file_path)
|
||
|
|
print(response)
|
||
|
|
|
||
|
|
|
||
|
|
if __name__ == "__main__":
|
||
|
|
main()
|