PythonのFlaskを使い、Whisperを呼び出して文字起こしをするAPIを実装します。
Ubuntu 20.04のコンピュート・インスタンスの方がWhisperの処理が速いため、Ubuntu上にAPIサーバーを実装します。
最終的にAutonomous Databaseに作成するAPEXアプリケーションから呼び出すため、ホスト名がDNSに登録されIPアドレスの解決ができること、HTTPS化がされていることが必要になります。
サーバー証明書は、Let's Encryptより発行します。
DNSへのホスト名とIPアドレスの登録は手順に含みません。hostコマンドの実行で、IPアドレスが返ってくるような状態になっていることが前提です。
host ホスト名
ubuntu@mywhisper2:~$ host ホスト名
ホスト名 has address ***.***.***.***
ubuntu@mywhisper2:~$
あらかじめ、パブリック・ネットワークのイングレス・ルールにて、ポート80（http）と443（https）への通信を許可しておきます。
mkdir certs audio
ubuntu@mywhisper2:~$ cd
ubuntu@mywhisper2:~$ mkdir certs audio
ubuntu@mywhisper2:~$
sudo apt install firewalld certbot
ubuntu@mywhisper2:~$ sudo apt install firewalld certbot
Reading package lists... Done
Building dependency tree
Reading state information... Done
The following additional packages will be installed:
[中略]
Processing triggers for systemd (245.4-4ubuntu3.19) ...
Processing triggers for man-db (2.9.1-1) ...
Processing triggers for dbus (1.12.16-2ubuntu2.3) ...
Processing triggers for libc-bin (2.31-0ubuntu9.9) ...
ubuntu@mywhisper2:~$
Flaskをインストールします。
ubuntu@mywhisper2:~$ pip install flask
Collecting flask
Downloading Flask-2.2.2-py3-none-any.whl (101 kB)
|████████████████████████████████| 101 kB 13.2 MB/s
Collecting Werkzeug>=2.2.2
Downloading Werkzeug-2.2.2-py3-none-any.whl (232 kB)
|████████████████████████████████| 232 kB 76.9 MB/s
Collecting click>=8.0
Downloading click-8.1.3-py3-none-any.whl (96 kB)
|████████████████████████████████| 96 kB 10.0 MB/s
Collecting itsdangerous>=2.0
Downloading itsdangerous-2.1.2-py3-none-any.whl (15 kB)
Requirement already satisfied: importlib-metadata>=3.6.0; python_version < "3.10" in /usr/local/lib/python3.8/dist-packages (from flask) (5.1.0)
Collecting Jinja2>=3.0
Downloading Jinja2-3.1.2-py3-none-any.whl (133 kB)
|████████████████████████████████| 133 kB 80.5 MB/s
Collecting MarkupSafe>=2.1.1
Downloading MarkupSafe-2.1.2-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl (26 kB)
Requirement already satisfied: zipp>=0.5 in /usr/lib/python3/dist-packages (from importlib-metadata>=3.6.0; python_version < "3.10"->flask) (1.0.0)
Installing collected packages: MarkupSafe, Werkzeug, click, itsdangerous, Jinja2, flask
Successfully installed Jinja2-3.1.2 MarkupSafe-2.1.2 Werkzeug-2.2.2 click-8.1.3 flask-2.2.2 itsdangerous-2.1.2
ubuntu@mywhisper2:~$
firewalldを構成します。
sudo firewall-cmd --add-service=http
sudo firewall-cmd --add-service=https
sudo firewall-cmd --add-forward-port=port=443:proto=tcp:toport=8443
sudo firewall-cmd --runtime-to-permanent
sudo firewall-cmd --reload
sudo firewall-cmd --list-all
ubuntu@mywhisper2:~$ sudo firewall-cmd --add-service=http
success
ubuntu@mywhisper2:~$ sudo firewall-cmd --add-service=https
success
ubuntu@mywhisper2:~$ sudo firewall-cmd --add-forward-port=port=443:proto=tcp:toport=8443
success
ubuntu@mywhisper2:~$ sudo firewall-cmd --runtime-to-permanent
success
ubuntu@mywhisper2:~$ sudo firewall-cmd --reload
success
ubuntu@mywhisper2:~$ sudo firewall-cmd --list-all
public
target: default
icmp-block-inversion: no
interfaces:
sources:
services: dhcpv6-client http https ssh
ports:
protocols:
masquerade: no
forward-ports: port=443:proto=tcp:toport=8443:toaddr=
source-ports:
icmp-blocks:
rich rules:
ubuntu@mywhisper2:~$
ubuntu@mywhisper2:~$ sudo certbot certonly --standalone
/usr/lib/python3/dist-packages/requests/__init__.py:89: RequestsDependencyWarning: urllib3 (1.26.13) or chardet (3.0.4) doesn't match a supported version!
warnings.warn("urllib3 ({}) or chardet ({}) doesn't match a supported "
Saving debug log to /var/log/letsencrypt/letsencrypt.log
Plugins selected: Authenticator standalone, Installer None
Enter email address (used for urgent renewal and security notices) (Enter 'c' to
cancel): 申請者のメール・アドレス
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Please read the Terms of Service at
https://letsencrypt.org/documents/LE-SA-v1.3-September-21-2022.pdf. You must
agree in order to register with the ACME server at
https://acme-v02.api.letsencrypt.org/directory
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
(A)gree/(C)ancel: A
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Would you be willing to share your email address with the Electronic Frontier
Foundation, a founding partner of the Let's Encrypt project and the non-profit
organization that develops Certbot? We'd like to send you email about our work
encrypting the web, EFF news, campaigns, and ways to support digital freedom.
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
(Y)es/(N)o: N
Please enter in your domain name(s) (comma and/or space separated) (Enter 'c'
to cancel): サーバー証明書を発行するホスト名
Obtaining a new certificate
Performing the following challenges:
http-01 challenge for ホスト名
Waiting for verification...
Cleaning up challenges
IMPORTANT NOTES:
- Congratulations! Your certificate and chain have been saved at:
/etc/letsencrypt/live/ホスト名/fullchain.pem
Your key file has been saved at:
/etc/letsencrypt/live/ホスト名/privkey.pem
Your cert will expire on 2023-04-23. To obtain a new or tweaked
version of this certificate in the future, simply run certbot
again. To non-interactively renew *all* of your certificates, run
"certbot renew"
- If you like Certbot, please consider supporting our work by:
Donating to ISRG / Let's Encrypt: https://letsencrypt.org/donate
Donating to EFF: https://eff.org/donate-le
ubuntu@mywhisper2:~$
sudo cp /etc/letsencrypt/live/ホスト名/privkey.pem certs
sudo cp /etc/letsencrypt/live/ホスト名/fullchain.pem certs
sudo chown ubuntu certs/*
sudo chmod 400 certs/*
ubuntu@mywhisper2:~$ sudo cp /etc/letsencrypt/live/ホスト名/privkey.pem certs
ubuntu@mywhisper2:~$ sudo cp /etc/letsencrypt/live/ホスト名/fullchain.pem certs
ubuntu@mywhisper2:~$ ls certs
fullchain.pem privkey.pem
ubuntu@mywhisper2:~$ sudo chown ubuntu certs/*
ubuntu@mywhisper2:~$ sudo chmod 400 certs/*
ubuntu@mywhisper2:~$ python whisper-server.py
/usr/lib/python3/dist-packages/requests/__init__.py:89: RequestsDependencyWarning: urllib3 (1.26.13) or chardet (3.0.4) doesn't match a supported version!
warnings.warn("urllib3 ({}) or chardet ({}) doesn't match a supported "
* Serving Flask app 'whisper-server'
* Debug mode: on
WARNING: This is a development server. Do not use it in a production deployment. Use a production WSGI server instead.
* Running on all addresses (0.0.0.0)
* Running on https://127.0.0.1:8443
* Running on https://10.0.0.131:8443
Press CTRL+C to quit
* Restarting with stat
/usr/lib/python3/dist-packages/requests/__init__.py:89: RequestsDependencyWarning: urllib3 (1.26.13) or chardet (3.0.4) doesn't match a supported version!
warnings.warn("urllib3 ({}) or chardet ({}) doesn't match a supported "
* Debugger is active!
* Debugger PIN: 197-750-656
以上で、APIの呼び出しによる文字起こしができるようになりました。
% curl -X POST -F 'file=@test.m4a' https://ホスト名/transcribe
{
"language": "ja",
"segments": [
{
"avg_logprob": -0.39915904131802643,
"compression_ratio": 1.065217391304348,
"end": 10.6,
"id": 0,
"no_speech_prob": 0.02103608287870884,
"seek": 0,
"start": 0.0,
"temperature": 0.0,
"text": "\u3053\u3093\u306b\u3061\u306f\u521d\u3081\u3066Whipper\u3092\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u3057\u3066\u307f\u307e\u3057\u305f \u3053\u308c\u3067\u8a66\u3057\u3066\u307f\u307e\u3059",
"tokens": [
50364,
38088,
28727,
38975,
2471,
15124,
5998,
8040,
4824,
40498,
35307,
8822,
11362,
12072,
33732,
2474,
22099,
8822,
11362,
5368,
50894
]
}
],
"text": "\u3053\u3093\u306b\u3061\u306f\u521d\u3081\u3066Whipper\u3092\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u3057\u3066\u307f\u307e\u3057\u305f \u3053\u308c\u3067\u8a66\u3057\u3066\u307f\u307e\u3059"
}
%
