教程：使用 OpenSearch Service 和 OpenSearch 控制面板可视化客户支持呼叫

将以下 Python 3 示例代码复制并粘贴到名为 call-center.py 的新文件中：

import boto3
import datetime
import json
import requests
from requests_aws4auth import AWS4Auth
import time
import urllib.request

# Variables to update
audio_file_name = '' # For example, 000001.mp3
bucket_name = '' # For example, my-transcribe-test
domain = '' # For example, https://search-my-transcribe-test-12345.us-west-2.es.amazonaws.com
index = 'support-calls'
type = '_doc'
region = 'us-west-2'

# Upload audio file to S3.
s3_client = boto3.client('s3')

audio_file = open(audio_file_name, 'rb')

print('Uploading ' + audio_file_name + '...')
response = s3_client.put_object(
    Body=audio_file,
    Bucket=bucket_name,
    Key=audio_file_name
)

# # Build the URL to the audio file on S3.
# # Only for the us-east-1 region.
# mp3_uri = 'https://' + bucket_name + '.s3.amazonaws.com/' + audio_file_name

# Get the necessary details and build the URL to the audio file on S3.
# For all other regions.
response = s3_client.get_bucket_location(
    Bucket=bucket_name
)
bucket_region = response['LocationConstraint']
mp3_uri = 'https://' + bucket_name + '.s3-' + bucket_region + '.amazonaws.com/' + audio_file_name

# Start transcription job.
transcribe_client = boto3.client('transcribe')

print('Starting transcription job...')
response = transcribe_client.start_transcription_job(
    TranscriptionJobName=audio_file_name,
    LanguageCode='en-US',
    MediaFormat='mp3',
    Media={
        'MediaFileUri': mp3_uri
    },
    Settings={
        'ShowSpeakerLabels': True,
        'MaxSpeakerLabels': 2 # assumes two people on a phone call
    }
)

# Wait for the transcription job to finish.
print('Waiting for job to complete...')
while True:
    response = transcribe_client.get_transcription_job(TranscriptionJobName=audio_file_name)
    if response['TranscriptionJob']['TranscriptionJobStatus'] in ['COMPLETED', 'FAILED']:
        break
    else:
        print('Still waiting...')
    time.sleep(10)

transcript_uri = response['TranscriptionJob']['Transcript']['TranscriptFileUri']

# Open the JSON file, read it, and get the transcript.
response = urllib.request.urlopen(transcript_uri)
raw_json = response.read()
loaded_json = json.loads(raw_json)
transcript = loaded_json['results']['transcripts'][0]['transcript']

# Send transcript to Comprehend for key phrases and sentiment.
comprehend_client = boto3.client('comprehend')

# If necessary, trim the transcript.
# If the transcript is more than 5 KB, the Comprehend calls fail.
if len(transcript) > 5000:
    trimmed_transcript = transcript[:5000]
else:
    trimmed_transcript = transcript

print('Detecting key phrases...')
response = comprehend_client.detect_key_phrases(
    Text=trimmed_transcript,
    LanguageCode='en'
)

keywords = []
for keyword in response['KeyPhrases']:
    keywords.append(keyword['Text'])

print('Detecting sentiment...')
response = comprehend_client.detect_sentiment(
    Text=trimmed_transcript,
    LanguageCode='en'
)

sentiment = response['Sentiment']

# Build the Amazon OpenSearch Service URL.
id = audio_file_name.strip('.mp3')
url = domain + '/' + index + '/' + type + '/' + id

# Create the JSON document.
json_document = {'transcript': transcript, 'keywords': keywords, 'sentiment': sentiment, 'timestamp': datetime.datetime.now().isoformat()}

# Provide all details necessary to sign the indexing request.
credentials = boto3.Session().get_credentials()
awsauth = AWS4Auth(credentials.access_key, credentials.secret_key, region, 'opensearchservice', session_token=credentials.token)

# Index the document.
print('Indexing document...')
response = requests.put(url, auth=awsauth, json=json_document, headers=headers)

print(response)
print(response.json())

更新最初的六个变量。

使用以下命令安装所需的程序包：

pip install boto3
pip install requests
pip install requests_aws4auth

将您的 MP3 与 call-center.py 放在同一目录中并运行脚本。示例输出如下：

$ python call-center.py
Uploading 000001.mp3...
Starting transcription job...
Waiting for job to complete...
Still waiting...
Still waiting...
Still waiting...
Still waiting...
Still waiting...
Still waiting...
Still waiting...
Detecting key phrases...
Detecting sentiment...
Indexing document...
<Response [201]>
{u'_type': u'call', u'_seq_no': 0, u'_shards': {u'successful': 1, u'failed': 0, u'total': 2}, u'_index': u'support-calls4', u'_version': 1, u'_primary_term': 1, u'result': u'created', u'_id': u'000001'}

创建一个名为 bulk-helper.py 的文件：

import boto3
from opensearchpy import OpenSearch, RequestsHttpConnection
import json
from requests_aws4auth import AWS4Auth

host = '' # For example, my-test-domain.us-west-2.es.amazonaws.com
region = '' # For example, us-west-2
service = 'es'

bulk_file = open('sample-calls.bulk', 'r').read()

credentials = boto3.Session().get_credentials()
awsauth = AWS4Auth(credentials.access_key, credentials.secret_key, region, service, session_token=credentials.token)

search = OpenSearch(
    hosts = [{'host': host, 'port': 443}],
    http_auth = awsauth,
    use_ssl = True,
    verify_certs = True,
    connection_class = RequestsHttpConnection
)

response = search.bulk(bulk_file)
print(json.dumps(response, indent=2, sort_keys=True))

为 host 和 region 更新最初的两个变量。

使用以下命令安装所需的程序包：

pip install opensearch-py

下载并解压缩 sample-calls.zip。

将 sample-calls.bulk 与 bulk-helper.py 放在同一目录中并运行帮助程序。示例输出如下：

$ python bulk-helper.py
{
  "errors": false,
  "items": [
    {
      "index": {
        "_id": "1",
        "_index": "support-calls",
        "_primary_term": 1,
        "_seq_no": 42,
        "_shards": {
          "failed": 0,
          "successful": 1,
          "total": 2
        },
        "_type": "_doc",
        "_version": 9,
        "result": "updated",
        "status": 200
      }
    },
    ...
  ],
  "took": 27
}

导航到 https://search-domain.region.es.amazonaws.com/_dashboards。

在使用 OpenSearch 控制面板之前，您需要索引模式。控制面板使用索引模式来将分析范围缩小到一个或多个索引。要匹配 call-center.py 创建的 support-calls 索引，转到堆栈管理、索引模式，并定义索引模式support*，然后选择下一步。

对于 Time Filter field name (时间筛选字段名称)，选择 timestamp (时间戳)。

现在，您可以开始创建可视化了。选择 Visualize (可视化)，然后添加新的可视化。

选择饼图和 support* 索引模式。

默认可视化是基本的，因此请选择 Split Slices (拆分切片) 来创建更有趣的可视化。

对于 Aggregation，选择 Terms。对于 Field (字段)，请选择 sentiment.keyword。然后选择 Apply changes (应用更改) 和 Save (保存)。

返回到 Visualize (可视化) 页面，然后添加其他可视化。这次请选择水平条形图。

选择 Split Series (拆分序列)。

对于 Aggregation，选择 Terms。对于 Field (字段)，请选择 keywords.keyword，并将 Size (大小) 更改为 20。然后选择 Apply Changes (应用更改) 和 Save (保存)。

返回到 Visualize (可视化) 页面并添加一个最终的可视化、一个垂直条形图。

选择 Split Series (拆分序列)。对于 Aggregation (聚合)，请选择 Date Histogram (日期直方图)。对于 Field (字段)，请选择 timestamp (时间戳) 并将 Interval (间隔) 更改为 Daily (每日)。

选择 Metrics & Axes (指标和轴)，并将 Mode (模式) 更改为 normal (正常)。

选择 Apply Changes (应用更改) 和 Save (保存)。

现在，您有三个可视化内容，可以将它们添加到控制面板。选择 Dashboard (控制面板)，创建一个控制面板，并添加您的可视化内容。