Official Python <=2.6 Helper Library for the Clarify API. For Python 2.7/3.x use When in doubt, check out the full documentation here: or contact us at
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Failed to load latest commit information.


Clarify Python <=2.6 Helper Library

For Python 2.7 and 3.x, use

Python 2.6 helper library for the Clarify API

  • Free software: MIT license


$ pip install clarify_python_2

You may need to use sudo if you don't have permission to install.


If you are running an older version of the python helper library, please upgrade.

$ pip install --upgrade clarify_python_2

You may need to use sudo if you don't have permission to upgrade.

Quickstart Guide

Getting Started

This quickstart demonstrates a simple way to get started using the Clarify API. Following these steps, it should take you no more than 5-10 minutes to have a fully functional search for your audio.

Configuring Your Environment

While you can use any programming language you choose, we provide helping libraries in a few to get you started. In Python, you simply include the Clarify file from the python_2 module, and initialize the environment with your API key:

from clarify_python_2 import clarify

clarify.set_key('my api key')

Loading Audio

Once you've initialied the environment with your API key, you load a file like this:

clarify.create_bundle(name='test bundle', media_url='')

Naming the bundle is optional.

Here are some audio files you can use for testing:

Hint: You don't have to download these files. Instead you can pass us these urls via the create_bundle() method shown above.

Searching Audio

To search, we'll use the search() function. If you uploaded the Wizard of Oz audio clip, you can search for "dorothy":'dorothy')

Then you can process and interact the results however you wish. The code below simply shows the resulting bundle id, bundle name, and the start/end offsets for each occurrence of the search terms. This assumes that the audio clip has been indexed by the time you search. If it hasn't, wait and try again in a few minutes.

result ='dorothy')
results = result['item_results']
items = result['_links']['items']

index = 0
for item in items:
    bundle = clarify.get_bundle(item['href'])

    print bundle['name']

    search_hits = results[index]['term_results'][0]['matches'][0]['hits']
    for search_hit in search_hits:
        print str(search_hit['start']) + ' -- ' + str(search_hit['end'])


And here are the results using the Wizard of Oz clip we loaded.

dorothy and her friends
2.35 -- 2.59
172.49 -- 172.83
224.82 -- 225.08
271.49 -- 271.8
329.1 -- 329.31
480.45 -- 480.92

Putting it all Together

From here, we can visualize our search results with the included audio player. The player should work with no additional configuration, but the bulk of the logic is here:

import json

result ='dorothy')
search_terms = json.dumps(result['search_terms'])
item_results = json.dumps(result['item_results'])

bundleref = result['_links']['items'][0]['href']
bundle = clarify.get_bundle(bundleref)
tracksref = bundle['_links']['clarify:tracks']['href']
tracks = clarify.get_track_list(tracksref)['tracks']
mediaURL = tracks[0]['media_url']

History (Change Log)



See TODO.rst