Content Moderation

1
import requests
2
import time
3
4
base_url = "https://api.assemblyai.com"
5
6
headers = {
7
    "authorization": "<YOUR_API_KEY>"
8
}
9
10
with open("./local_file.mp3", "rb") as f:
11
    response = requests.post(base_url + "/v2/upload",
12
                            headers=headers,
13
                            data=f)
14
15
upload_url = response.json()["upload_url"]
16
17
data = {
18
    "audio_url": upload_url, # You can also use a URL to an audio or video file on the web
19
    "speech_models": ["universal-3-pro", "universal-2"],
20
    "language_detection": True,
21
    "content_safety": True
22
}
23
24
url = base_url + "/v2/transcript"
25
response = requests.post(url, json=data, headers=headers)
26
27
transcript_id = response.json()['id']
28
polling_endpoint = base_url + "/v2/transcript/" + transcript_id
29
30
print(f"Transcript ID: {transcript_id}")
31
32
while True:
33
    transcription_result = requests.get(polling_endpoint, headers=headers).json()
34
35
    if transcription_result['status'] == 'completed':
36
37
        for result in transcription_result['content_safety_labels']['results']:
38
            print(result['text'])
39
            print(f"Timestamp: {result['timestamp']['start']} - {result['timestamp']['end']}")
40
41
            # Get category, confidence, and severity.
42
            for label in result['labels']:
43
                print(f"{label['label']} - {label['confidence']} - {label['severity']}")  # content safety category
44
45
        # Get the confidence of the most common labels in relation to the entire audio file.
46
        for label, confidence in transcription_result['content_safety_labels']['summary'].items():
47
            print(f"{confidence * 100}% confident that the audio contains {label}")
48
49
        # Get the overall severity of the most common labels in relation to the entire audio file.
50
        for label, severity_confidence in transcription_result['content_safety_labels']['severity_score_summary'].items():
51
            print(f"{severity_confidence['low'] * 100}% confident that the audio contains low-severity {label}")
52
            print(f"{severity_confidence['medium'] * 100}% confident that the audio contains medium-severity {label}")
53
            print(f"{severity_confidence['high'] * 100}% confident that the audio contains high-severity {label}")
54
        break
55
    elif transcription_result['status'] == 'error':
56
        raise RuntimeError(f"Transcription failed: {transcription_result['error']}")
57
    else:
58
        time.sleep(3)

1
# Setting the content safety confidence threshold to 60%.
2
data = {
3
    "audio_url": upload_url,
4
    "content_safety": True,
5
    "content_safety_confidence": 60
6
}