We annotate
audio for you

Scale quickly with high quality data

audio labels by professionals

We are flexible.

Same same but different. Each audio and machine learning use case is unique and therefore needs customised data. Speech, sentiment, tones, notes, instruments, events or scenes… we have supported a variety of different projects with various forms and granularity of data and  annotations. In order to guarantee.

We are quick.

Our workforce is ready on-demand. You can upload your data manually or via API and we return high-accuracy labels within the shortest possible time horizon. With intelligent preselection algorithms we are able to select the best annotation experts for your task. We have annotated vast amounts of audio and caN react to traffic peaks.

We can scale.

No dataset is too large, no dataset too complicated. We can scale our capacities and develop solutions to fulfil your specific needs. We are providing an enterprise solution for the growing demand of annotated audio data. As a scalable, high quality data provider we help pushing your application to the next level.

We know audio.

Garbage in, garbage out. High quality labels are the gold standard for every machine learning application. That’s why we rely on highly skilled audio professionals. Our team is individually trained on specific tasks to ensure high labelling accuracies. We have encountered the pitfalls of audio machine learning projects and can assist you in the setup and preparation. 

leverage audio intelligence

Audio data contains a plethora of hidden information. We have taught ourselves to rely on that information in daily life but we are just at the beginning of leveraging this information for intelligent applications. Automated audio analysis can facilitate processes, optimise workflows or detect specific situations. While video applications have lead the way, many AI challenges can only be tackled with auditive sense. With audio-annotate you can start to unleash the power of sound analysis and built your project around a reliable and high-quality data basis.

specialized tools

Audio data is a special breed. We have built our own tools to leverage the human input to the max. Our annotation process relies on constant feedback from our team as well as our customers. In order to improve our annotation results, we have adopted preselection and intelligent scanning methods to advise and consult our customers. We are not looking for a one-step solution to generalise all needs. Instead, we work towards high quality audio labels even for the most sophisticated need.

categories and pricing

  • TYPE
    APPROXIMATE COST [per minute audio]
  • Tagging / Category

    $20 / 1000 annotations

  • Event / Region / Bounding box
    $30 / 1000 annotations
  • Speech Emotion / Sentiment
    $35 / 1000 annotations
  • Music Segments / Boundaries
    $25 / 1000 annotations
  • Music Genre
    $20 / 1000 annotations
  • Musical instrument
    $35 / 1000 annotations

how works

Initial Workshop:

We are discussing your use case in order to understand how we can best tackle the challenge together.  

Data exploration:

By sharing a snapshot of your data, we assess our labeling strategy  and give you a time & pricing indication.

Project setup:

Based on your guidelines, we create references and qualify our team for your data structures.

Test and refinement:

We‘ provide annotations on a small test set of audio and discuss your feedback to improve the process.

Your high quality dataset,

ready for production: We move towards production mode and process the full amount of data. However, there’s always the possibility to re-iterate over the previous steps and optimise our workflow or data specifications.

contact us

    I have read the privacy policy and accept it.

    Datenschutz | Impressum | realization by Andersen Webworks