Why Stereo Recording is the Best Choice for Contact Centers

Traditionally, contact center calls are recorded in mono. This means that one channel is used to record all parties on a call. Stereo recording is different from mono recording in that it records each party through a unique channel. Typically this is two channels, but it can be many more.

Depending on which recording method you use, the resulting audio will be different. Take for example, you listen back to a call recording between two participants using a pair of headphones. If recorded in mono, both speakers can be heard equally through the left and right channels. Yet if recorded in stereo, each speaker is played back through a single channel – one through the left, and the other the right.

Summary

• Mono recording: each participant is recorded on, and played back via, the same channel.

• Stereo recording: each participant is recorded on, and played back via, their own unique channel.

In conversation there is usually some degree of overtalk, that is, people speaking over one another. Throw background noise into the mix, such as traffic or office hubbub, and a conversation can easily turn into a wall of noise. Overtalk and background noise make it difficult to tell speakers apart, which consequently cause transcription technologies, like automatic speech recognition (ASR), to produce many errors.

Recording each participant on a separate channel produces higher-fidelity audio: a clearer, wider and more natural sound. Each participant can be identified and isolated from one another. This makes it much easier to hear what each participant is saying, especially where there is overtalk or background noise from other participants. In comparison to a mono recording, because all participants are recorded on a single channel, any background noise or overtalk will muddy the entire recording. Telling participants apart in this case becomes very difficult.

As a result of more precise audio, Speech-To-Text (STT) transcription is more accurate. Typically, multi-channel recordings lead to a 10% increase in transcription accuracy compared to mono recording. A higher transcription accuracy allows you to unleash the full power of AI, including Natural Language Processing (NLP), for analyzing calls for training and compliance purposes.

In short, stereo recording leads to higher-quality audio, more accurate transcription, and greater opportunities for analytics.

Traditional call analysis relies on administrators taking random samples of call recordings and listening to them to look for issues of compliance, or to identify gaps in training. Random call sampling is an incredibly time consuming process – but it doesn’t have to be this way.

Stereo recording is the bedrock your contact center needs to extract greater business value from your call recordings. Higher-quality audio, and thus a higher transcription accuracy, magnifies the abilities of your AI to monitor all calls and participants, and to flag areas of concern, automatically. AI reduces the need for human monitoring of random audio samples. This means less time listening to calls, giving more time for when you do need to review recordings manually. Then, quality monitoring teams can focus on areas of concern, and bypass calls with no added value.

The ability to track the right data automatically is invaluable. With higher fidelity audio, your contact center administrators can focus on areas of interest that NLP has highlighted, ensuring greater productivity and more effective quality management (QM).