Storacall AudioSearch Benefits

  • Fast analysis of call recordings for valuable agent and customer interaction intelligence
  • Contact centres can use AudioSearch to look for issues such as long hold times, adherence to scripts, general process issues and automation issues.
  • Enables the proactive management of agents with better coaching, to give improvements in quality, reduced silence time and reduced hold times
  • Provides vital information for management teams to improve operational efficiency, risk management and revenue generation
  • Speed and Scalability: Instantly index and search audio content from your desktop or laptop computer without the need for integration. Audio can be searched at 360 million times faster than real-time.
  • Easy to Deploy: AudioSearch can be easily installed on a laptop or desktop in a matter of minutes. The simple interface makes it easy for non-technical operators to quickly become acclimated with the application.
  • Accurate Threat Detection: Works across all types of spoken audio and provides a fast and efficient means of organizing and sharing search results.
  • Portability: Provides a full set of audio review features that can be installed on a laptop, which is ideal for remote operations. As a stand-alone application, it can accommodate anywhere from 60,000 to 100,000 media files at top performance.
  • Collaboration: Supports a variety of import and export functions so work can be shared between one or more users. Reviewers can easily organise files into multiple sets, flexible export options allow sending both the audio files and their meta-data in different formats.

The Technology behind AudioSearch

AudioSearch is powered by Nexidia’s award-winning, Phonetic Search Engine technology, which searches audio and video using phonemes the smallest unit of human speech. Media files are phonetically indexed—broken down into phonemes which can then be searched for the most accurate, relevant results. This phonetic approach supports almost all generally available audio qualities and audio variances such as a speaker’s language, accent, dialect, gender and age.

This phonetic solution vastly accelerates the audio mining process through “automated listening,” which systematically ingests and identifies content within voice recordings. AudioSearch delivers timely identification of threats and trends contained within these recordings.


  • Automatically sorts recordings by respective languages, allowing users to quickly conduct searches within each native language group. Also detects and classifies a wide range of audio variances, such as a speaker's accent, dialect, gender and age.
  • Allows for complex, phonetic-based search using BOOLEAN logic (e.g. AND, OR, NOT) and even time-based proximity logic to a query.
  • The Pronunciation Optimisation feature allows users to do test searches and then generate a new search term/phrase that most closely represents the best results. It is also possible to search by an example clip where the user can identify the specific segment of audio containing the desired term and then use it as the search string.
  • Allows users to take examples of speech (a clip or an entire file) from a speaker of interest to create a model for that individual that can then be used to search for that speaker in other files.
  • Presents results of search terms in a table organised by how accurate the search term was to the audio found, this allows you to identify the most accurate results and also using optimisation you can ‘train’ the system to ensure you get the right results quickly and efficiently.
  • Enables clips to be redacted from audio files. Redacted portions of the file are not playable or searchable to protect sensitive information.
  • Increases performance due to the 64-bit environment
  • Warns users if pre-requisites and/or hardware configurations do not meet recommendations when installing

Our sales team will be happy to advice on the best solution for your organisation call +44(0)1932 710710 or Click Here to Request a Brochure