audionamix.js

Audionamix Audio SDK Node.js

How to use this client

You have to specifiy your access key and secret environment variables to use this client. See http://audionamix.com/ for more info.

$ export AUDIONAMIX_ACCESS_KEY=xxxxxx
$ export AUDIONAMIX_SECRET=xxxxxxxxxxxxxxxxxxxxxxxxxx

A simple Workflow

Upload a wave file

$ node upload sample.wav 
Uploading... sample.wav
upload { channel_number: null,
  duration: -1,
  file: 'https://.../sample.wav',
  id: 314329,
  ip: 'xxx.xx.xx.xxx',
  is_valid: true,
  original_file_type: null,
  resource_uri: '/api/v1/audiofile/314329/',
  upload_duration: '0:00:03.047420',
  valid: true }

Start the separation process passing the audiofile id in the response above.

$ node separation 314329
separation { extracted_file_id: 314330,
  id: 120571,
  input_file_id: 314329,
  resource_uri: '/api/v1/separation/120571/',
  status: 1 }

Query the separatation status passing the separation id in the response above:

$ node status 120571
status { extracted_file_id: 314330,
  id: 120571,
  input_file_id: 314329,
  resource_uri: '/api/v1/separation/120571/',
  status: 50 }

When the status value is 100, download the separated wave output file using the input_file_id in the previous status response:

$ node download 314330
download ./sample_extracted_314330.wav

Enhanced Workflow

Upload the audio wave file

$ node upload sample2.wav 
Uploading... sample2.wav
upload { channel_number: null,
  duration: -1,
  file: 'https://xxxx/sample2_fanTKcD.wav',
  id: 316517,
  ip: 'xx.xx.xx.xxx',
  is_valid: true,
  original_file_type: null,
  resource_uri: '/api/v1/audiofile/316517/',
  upload_duration: '0:00:02.741144',
  valid: true }

Launch a preanalysis with pitch algorithm for voice pitch detection

$ node preanalysis 316517 pitch
Algorithm will be pitch
preanalysis { id: 29446,
  is_finished: false,
  resource_uri: '/api/v1/preanalysis/29446/',
  status: 0 }

Launch a preanalysis with csnt algorithm for voice consonants detection

$ node preanalysis 316517 csnt
Algorithm will be csnt

preanalysis { id: 29447,
  is_finished: false,
  resource_uri: '/api/v1/preanalysis/29447/',
  status: 0 }

Check the preanalysis status for these analysis

$ node status_preanalysis 29446 5
check status every 5 secs.

Fri Oct 07 2016 16:25:01 GMT+0200 (CEST) 
To stop press Ctrl-C
status
 { id: 29446,
  is_finished: true,
  resource_uri: '/api/v1/preanalysis/29446/',
  status: 100 }

and

$ node status_preanalysis 29447 5
check status every 5 secs.

Fri Oct 07 2016 16:25:27 GMT+0200 (CEST) 
To stop press Ctrl-C
status
 { id: 29447,
  is_finished: true,
  resource_uri: '/api/v1/preanalysis/29447/',
  status: 100 }

Retrieve preanalysis annotations

$ node annotation 29446
annotation ./annotation_29446.json
$ node annotation 29447
annotation ./annotation_29447.json

Merge the two json files in order to keep the keys (not overlapping) in it like

var FileUtil=require('../lib/fileutil');
var annotation1 = FileUtil.readFileSync(annotationFile1,true);
var annotation2 = FileUtil.readFileSync(annotationFile2,true);
for(var key in annotation2) {
    if( typeof(annotation1[key])!='undefined' ) annotation1[key] = annotation2[key];
}
FileUtil.writeToFileSync(configurationFile, JSON.stringify(annotation1));

Upload the merged annotations file

$ node configuration annotation_merged.json 
Uploading... annotation_merged.json
configuration { big_dipper: true,
  breathiness: false,
  consonants_activity: '[]',
  extra_parameters_patch: '',
  high_quality: false,
  id: 334785,
  maximum_frequency: -1,
  minimum_frequency: -1,
  panning_vector: null,
  preference_name: null,
  processed_segments: null,
  rectangle_annotation: '',
  resource_uri: '/api/v1/configuration/334785/',
  reverb_time: 0,
  voice_activity: null }

Start the separation process for the file id obtained above and the configuration id of the previous json response

$ node separation 316517 334785
Configuration will be 334785
separation { configuration_id: 334785,
  extracted_file_id: 316525,
  id: 121318,
  input_file_id: 316517,
  resource_uri: '/api/v1/separation/121318/',
  status: 1 }

Check the separation status for this separation id

$ node status_separation 121318
check status every 5 secs.

Fri Oct 07 2016 16:38:12 GMT+0200 (CEST) 
To stop press Ctrl-C
status
 { configuration_id: 334785,
  extracted_file_id: 316525,
  id: 121318,
  input_file_id: 316517,
  resource_uri: '/api/v1/separation/121318/',
  status: 100 }

When the status is 100, download the separated wave file for the extrated file id specified in the response above

$ node download 316525
download ./sample_extracted_316525.wav

Fully Automated Enhanced Workflow

To run a fully automated enhanced worflow run the workflow example. The example will execute all steps from ehnanced workflow, performing both the preanalysis, downloading, merging and uploading the annotations files and downloading the extracted resulting file. To do this run command with a file identifier (useful for batch processing), an input wave file, the output path and the poll interval in seconds (optional):

Usage: worflow fileId filePath outputPath [poll_interval_seconds]
fileId		file identifier useful for batch processing of multiple files
filePath		file absolute path
outputPath	output files absolute path, defaults to script path

so we have

$ node workflow 1234 sample.wave out/ 5

The analysis files will be in the folder ./out/1234:

$ cd out/1234
$ ls -l
total 9432
-rw-r--r--  1 admin  staff   267082 21 Nov 15:46 annotation_37168_37169.json
-rw-r--r--  1 admin  staff       29 21 Nov 15:46 annotation_csnt_37169.json
-rw-r--r--  1 admin  staff   267058 21 Nov 15:46 annotation_pitch_37168.json
-rw-r--r--  1 admin  staff  4284126 21 Nov 15:46 sample_extracted_348787.wav