Analyze The Sentiment Of A Customer Call using LLM Gateway

1
import requests
2
import time
3
import json
4
5
API_KEY = "YOUR_API_KEY"
6
audio_file_path = "./meeting.mp3"
7
8
# ------------------------------------------
9
# Step 1: Upload the audio file
10
# ------------------------------------------
11
def upload_file(filename):
12
    with open(filename, "rb") as f:
13
        upload_url = "https://api.assemblyai.com/v2/upload"
14
        headers = {"authorization": API_KEY}
15
        response = requests.post(upload_url, headers=headers, data=f)
16
        response.raise_for_status()
17
        return response.json()["upload_url"]
18
19
audio_url = upload_file(audio_file_path)
20
print(f"Uploaded audio file. URL: {audio_url}")
21
22
# ------------------------------------------
23
# Step 2: Request transcription
24
# ------------------------------------------
25
transcript_request = requests.post(
26
    "https://api.assemblyai.com/v2/transcript",
27
    headers={"authorization": API_KEY, "content-type": "application/json"},
28
    json={"audio_url": audio_url},
29
)
30
31
transcript_id = transcript_request.json()["id"]
32
33
# Poll until completed
34
while True:
35
    polling_response = requests.get(
36
        f"https://api.assemblyai.com/v2/transcript/{transcript_id}",
37
        headers={"authorization": API_KEY},
38
    )
39
    status = polling_response.json()["status"]
40
41
    if status == "completed":
42
        transcript_text = polling_response.json()["text"]
43
        break
44
    elif status == "error":
45
        raise RuntimeError(f"Transcription failed: {polling_response.json()['error']}")
46
    else:
47
        print(f"Transcription status: {status}")
48
        time.sleep(3)
49
50
print("\nTranscription complete.\n")
51
52
# ------------------------------------------
53
# Step 3: Define questions
54
# ------------------------------------------
55
agent_context = "The agent is trying to get the customer to go through with the update to their car."
56
customer_context = "The customer is calling to check how much it would cost to update the map in his car."
57
58
answer_format = "<answer in one sentence> <reason in one sentence>"
59
60
questions = [
61
    {
62
        "question": "What was the overall sentiment of the call?",
63
        "context": customer_context,
64
        "answer_format": answer_format,
65
    },
66
    {
67
        "question": "What was the sentiment of the agent in this call?",
68
        "context": agent_context,
69
        "answer_format": answer_format,
70
    },
71
    {
72
        "question": "What was the sentiment of the customer in this call?",
73
        "context": customer_context,
74
        "answer_format": answer_format,
75
    },
76
    {
77
        "question": "What quote best demonstrates the customer's level of interest?",
78
        "context": customer_context,
79
        "answer_format": answer_format,
80
    },
81
    {
82
        "question": "Provide a quote from the agent that demonstrates their level of enthusiasm.",
83
        "context": agent_context,
84
        "answer_format": answer_format,
85
    },
86
]
87
88
# ------------------------------------------
89
# Step 4: Build prompt for the LLM
90
# ------------------------------------------
91
question_strs = []
92
for q in questions:
93
    q_str = f"Question: {q['question']}"
94
    if q.get("context"):
95
        q_str += f"\nContext: {q['context']}"
96
    if q.get("answer_format"):
97
        q_str += f"\nAnswer Format: {q['answer_format']}"
98
    question_strs.append(q_str)
99
100
questions_prompt = "\n\n".join(question_strs)
101
102
prompt = f"""
103
You are an expert at analyzing call transcripts.
104
Given the series of questions below, answer them accurately and concisely.
105
When context or answer format is provided, use it to guide your answers.
106
107
Transcript:
108
{transcript_text}
109
110
Questions:
111
{questions_prompt}
112
"""
113
114
# ------------------------------------------
115
# Step 5: Query the LLM Gateway
116
# ------------------------------------------
117
headers = {"authorization": API_KEY}
118
119
response = requests.post(
120
    "https://llm-gateway.assemblyai.com/v1/chat/completions",
121
    headers=headers,
122
    json={
123
        "model": "claude-sonnet-4-5-20250929",
124
        "messages": [{"role": "user", "content": prompt}],
125
        "max_tokens": 2000,
126
    },
127
)
128
129
response_json = response.json()
130
llm_output = response_json["choices"][0]["message"]["content"]
131
132
# ------------------------------------------
133
# Step 6: Parse and display the results
134
# ------------------------------------------
135
print("\n--- LLM Responses ---\n")
136
print(llm_output)

1
import requests
2
import time
3
import json
4
5
API_KEY = "YOUR_API_KEY"

1
audio_file_path = "./meeting.mp3"
2
# ------------------------------------------
3
# Step 1: Upload the audio file
4
# ------------------------------------------
5
def upload_file(filename):
6
    with open(filename, "rb") as f:
7
        upload_url = "https://api.assemblyai.com/v2/upload"
8
        headers = {"authorization": API_KEY}
9
        response = requests.post(upload_url, headers=headers, data=f)
10
        response.raise_for_status()
11
        return response.json()["upload_url"]
12
audio_url = upload_file(audio_file_path)
13
print(f"Uploaded audio file. URL: {audio_url}")
14
# ------------------------------------------
15
# Step 2: Request transcription
16
# ------------------------------------------
17
transcript_request = requests.post(
18
    "https://api.assemblyai.com/v2/transcript",
19
    headers={"authorization": API_KEY, "content-type": "application/json"},
20
    json={"audio_url": audio_url},
21
)
22
transcript_id = transcript_request.json()["id"]
23
# Poll until completed
24
while True:
25
    polling_response = requests.get(
26
        f"https://api.assemblyai.com/v2/transcript/{transcript_id}",
27
        headers={"authorization": API_KEY},
28
    )
29
    status = polling_response.json()["status"]
30
    if status == "completed":
31
        transcript_text = polling_response.json()["text"]
32
        break
33
    elif status == "error":
34
        raise RuntimeError(f"Transcription failed: {polling_response.json()['error']}")
35
    else:
36
        print(f"Transcription status: {status}")
37
        time.sleep(3)
38
print("\nTranscription complete.\n")

1
agent_context = "The agent is trying to get the customer to go through with the update to their car."
2
customer_context = "The customer is calling to check how much it would cost to update the map in his car."

1
answer_format = "<answer in one sentence> <reason in one sentence>"
2
questions = [
3
    {
4
        "question": "What was the overall sentiment of the call?",
5
        "context": customer_context,
6
        "answer_format": answer_format,
7
    },
8
    {
9
        "question": "What was the sentiment of the agent in this call?",
10
        "context": agent_context,
11
        "answer_format": answer_format,
12
    },
13
    {
14
        "question": "What was the sentiment of the customer in this call?",
15
        "context": customer_context,
16
        "answer_format": answer_format,
17
    },
18
    {
19
        "question": "What quote best demonstrates the customer's level of interest?",
20
        "context": customer_context,
21
        "answer_format": answer_format,
22
    },
23
    {
24
        "question": "Provide a quote from the agent that demonstrates their level of enthusiasm.",
25
        "context": agent_context,
26
        "answer_format": answer_format,
27
    },
28
]

1
# ------------------------------------------
2
# Step 4: Build prompt for the LLM
3
# ------------------------------------------
4
question_strs = []
5
for q in questions:
6
    q_str = f"Question: {q['question']}"
7
    if q.get("context"):
8
        q_str += f"\nContext: {q['context']}"
9
    if q.get("answer_format"):
10
        q_str += f"\nAnswer Format: {q['answer_format']}"
11
    question_strs.append(q_str)
12
questions_prompt = "\n\n".join(question_strs)
13
prompt = f"""
14
You are an expert at analyzing call transcripts.
15
Given the series of questions below, answer them accurately and concisely.
16
When context or answer format is provided, use it to guide your answers.
17
Transcript:
18
{transcript_text}
19
Questions:
20
{questions_prompt}
21
"""

1
# ------------------------------------------
2
# Step 5: Query the LLM Gateway
3
# ------------------------------------------
4
headers = {"authorization": API_KEY}
5
response = requests.post(
6
    "https://llm-gateway.assemblyai.com/v1/chat/completions",
7
    headers=headers,
8
    json={
9
        "model": "claude-sonnet-4-5-20250929",
10
        "messages": [{"role": "user", "content": prompt}],
11
        "max_tokens": 2000,
12
    },
13
)
14
response_json = response.json()
15
llm_output = response_json["choices"][0]["message"]["content"]
16
# ------------------------------------------
17
# Step 6: Parse and display the results
18
# ------------------------------------------
19
print("\n--- LLM Responses ---\n")
20
print(llm_output)