IBM Watson Speech to Text in Python gives 404 when using model parameter

Question

I am testing the use of IBM Watson Speech to Text with Python. I was able to successfully test transcribing an audio in English but when I put the model parameter to change the language model for my language, a 404 not found error appears. I have already looked at the IBM page several times that explains the use of the model parameter and I can't understand what's missing. Can anyone help?

My code:

from ibm_watson import SpeechToTextV1
from ibm_cloud_sdk_core.authenticators import IAMAuthenticator

api = IAMAuthenticator("my_credential")
speech_2_text = SpeechToTextV1(authenticator=api)

speech_2_text.set_service_url("https://api.us-south.speech-to-text.watson.cloud.ibm.com/instances/20a185d6-6953-4334-9cea-e9f5ebc2267d?model=fr-FR_BroadbandModel")

with open("test.mp3", "rb") as audio_file:
    result = speech_2_text.recognize(
    audio=audio_file,content_type="audio/mp3"
    ).get_result()

Error message:

ibm_cloud_sdk_core\base_service.py", line 224, 
in send raise ApiException(ibm_cloud_sdk_core.api_exception.ApiException: Error: Not Found, Code: 404

Answer 1

The model should be passed as part of the recognize method

speech_recognition_results = speech_to_text.recognize(
        audio=audio_file,
        content_type='audio/mp3',
        word_alternatives_threshold=0.9,
        model='fr-FR_BroadbandModel'
    ).get_result()

Pasting the complete code that worked me for your reference

import json
from os.path import join, dirname
from ibm_watson import SpeechToTextV1
from ibm_cloud_sdk_core.authenticators import IAMAuthenticator

authenticator = IAMAuthenticator('<API_KEY>')
speech_to_text = SpeechToTextV1(
    authenticator=authenticator
)

speech_to_text.set_service_url('<URL>')

with open(join(dirname(__file__), './.', 'audio-file2.mp3'),
               'rb') as audio_file:
    speech_recognition_results = speech_to_text.recognize(
        audio=audio_file,
        content_type='audio/mp3',
        word_alternatives_threshold=0.9,
        model='fr-FR_BroadbandModel'
    ).get_result()
print(json.dumps(speech_recognition_results, indent=2))