https://github.com/kiibo382/audio-text-sls

https://github.com/kiibo382/audio-text-sls

Last synced: about 1 month ago
JSON representation

• Host: GitHub
• URL: https://github.com/kiibo382/audio-text-sls
• Owner: kiibo382
• Created: 2021-03-02T07:53:16.000Z (over 3 years ago)
• Default Branch: main
• Last Pushed: 2022-05-15T08:43:17.000Z (over 2 years ago)
• Last Synced: 2024-04-18T02:15:31.944Z (7 months ago)
• Language: Python
• Size: 813 KB
• Stars: 0
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Serverless Application

## Architecture

![serverless](https://user-images.githubusercontent.com/64523345/110412110-01b29b00-80cf-11eb-9177-9d6af11d9333.png)

## Setup

### Installation

 - Node.js (version 14.40)

 - aws cli (including profile settings, etc.)

 - (optional) direnv

 - python3

  - boto3

### Common

```bash

echo 'export AWS_DEFAULT_PROFILE=Required profile' >> ~/.bash_profile

echo 'export AWS_SDK_LOAD_CONFIG=1' >> ~/.bash_profile

git clone https://github.com/kiibo382/audio-text-sls.git

```

## Usage

### Tutorial

```bash

cd sls-tutorial

npm install

# Then, adjust the settings (profile and bucket name) in serverless.yml and delete.sh, and execute the following.

npm run sls-deploy

```

### Local

```bash

cd local

npm install

# Add the following to `.aws/config`

[s3local]

region = ap-northeast-1

# Add the following to `.aws/credentials`

[s3local]

aws_access_key_id=S3RVER

aws_secret_access_key=S3RVER

# Then, adjust settings (profile and bucket name) in the .env file and execute the following.

npm run sls-offline

# Upload wav and json files to minio (S3 compatible object storage)

./upload.sh

# Access the following URL to check the operation.

# http://localhost:3000/records/{bucket_name}/{key}

# http://localhost:3000/results/{bucket_name}/{key}

```

### Production

```bash

cd cloud

npm install

npm run sls-deploy

```

After executing the command, please do the following.

1. after deployment, allow subscriptions you will receive from AWS SNS via email.

2. upload the wav file to the audio bucket

3. check if the path is received at the endpoint of the specified protocol (such as email).

4. access `https://{FQDN of the API Gateway}/{path received}` and confirm that the result is returned.

#### Note

The voice buckets to be processed are assumed to be two existing S3 buckets.

If you are creating a new voice bucket, please make sure that your `serverless.yml` is as follows

```yaml

fucntions:

  transcribe1:

    handler: transcribe.handler

    memorySize: 512

    events:

      - s3:

          bucket: ${env:RECORDS_BUCKET_NAME1}

          event: s3:ObjectCreated:*

          rules:

            - suffix: .wav

  transcribe2:

    handler: transcribe.handler

    memorySize: 512

    events:

      - s3:

          bucket: ${env:RECORDS_BUCKET_NAME2}

          event: s3:ObjectCreated:*

          rules:

            - suffix: .wav

```

To add a bucket, append the name of the S3 bucket to the `.env` file. (See example below.)

```jsx

RECORDS_BUCKET_NAME3=sls-sample-dev-records-bucket3

```

Also, add the following two items to `serverless.yaml`: provider.environment and functions.

```yaml

provider:

  environment:

    RECORDS_BUCKET_NAME3: ${env:RECORDS_BUCKET_NAME3}

.

.

.

functions:

  transcribe3:

    handler: transcribe.handler

    memorySize: 512

    events:

      - s3:

          bucket: ${env:RECORDS_BUCKET_NAME3}

          event: s3:ObjectCreated:*

          existing: true

          rules:

            - suffix: .wav

```

### Clean up

```bash

./delete.sh

```