{"id":9833,"date":"2023-02-13T22:24:22","date_gmt":"2023-02-13T18:24:22","guid":{"rendered":"https:\/\/krisp.ai\/blog\/?p=9833"},"modified":"2025-03-12T11:43:19","modified_gmt":"2025-03-12T07:43:19","slug":"speech-enhancement-review-krisp-use-case","status":"publish","type":"post","link":"https:\/\/krisp.ai\/blog\/speech-enhancement-review-krisp-use-case\/","title":{"rendered":"Speech Enhancement Review: Krisp Use Case"},"content":{"rendered":"

Imagine you have an important online meeting, and there is a lot of noise around you. Kids are playing, the dog is barking, the washing machine is running, a fan is turned on, there is construction happening nearby, and you need to join a call. More often than not, it is nearly impossible to stop the noise or find a quiet place. In such situations, we need special audio processing technology that can remove background noises to improve the quality of online meetings.\u00a0<\/span><\/p>\n

This is one of the best applications of speech enhancement. Here we will discuss speech enhancement technology, give some historical background, review existing approaches, cover the challenges surrounding real-time communication, and explore how Krisp\u2019s speech enhancement algorithm is an ideal solution.\u00a0\u00a0\u00a0\u00a0\u00a0<\/span><\/p>\n

First, let\u2019s define Speech Enhancement (SE). It improves the quality of a noisy speech signal by reducing or removing background noises (see Figure 1). The main goal is to improve the perceptual quality and intelligibility of speech distorted by noise.<\/span><\/p>\n

\"Figure<\/p>\n

Figure 1. Speech enhancement.<\/em><\/p>\n<\/div>\n

We sometimes find other terms used interchangeably with speech enhancement, such as noise cancellation (NC), noise reduction, noise suppression, and speech separation.\u00a0<\/span><\/p>\n

There are lots of applications for speech enhancement algorithms, including:<\/span><\/p>\n

    \n
  1. Voice communication, such as in conferencing apps, mobile phones, voice chats, and others. SE algorithms improve speech intelligibility for speakers in noisy environments, such as restaurants, offices, or crowded streets.\u00a0<\/span><\/li>\n
  2. Improving other types of audio processing algorithms by making them more noise-robust. For instance, we can apply speech enhancement prior to passing a signal to systems like speech recognition, speaker identification, speech emotion recognition, voice conversion, etc.\u00a0<\/span><\/li>\n
  3. Hearing aids. For those with hearing impairments, speech may be completely inaudible in noisy environments. Reducing noise increases intelligibility.<\/span><\/li>\n<\/ol>\n

    Traditional approaches<\/b><\/h2>\n

    The <\/span>first results<\/span><\/a> of research centered around speech enhancement were obtained in the 1970s. Traditional approaches were based on statistical assumptions and mathematical modeling of the problem. Their solutions also depend largely on the application, noise types, acoustic conditions, signal-to-noise ratio, and the number of available microphones (channels). Let\u2019s discuss them in more detail.<\/span><\/p>\n

    In general, we can divide speech enhancement algorithms into two types: <\/span>multi-channel<\/span><\/i> and <\/span>single-channel<\/span><\/i> (mono).\u00a0<\/span><\/p>\n

    The multi-channel case involves two or more microphones (channels). In this case, the extra channel(s) contain information on the noise signal and can help to reduce the noise signal in the primary channel. An example of such a method is <\/span>adaptive noise filtering<\/span><\/i><\/a>.\u00a0<\/span><\/p>\n

    This technique uses a reference signal from the auxiliary (secondary) microphone as an input to an adaptive digital filter, which estimates the noise in the primary signal and cancels it out (see Figure 2). Unlike a fixed filter, the adaptive filter automatically adjusts its <\/span>impulse response<\/span><\/i><\/a>. The adjustment is based on the error in the output. Therefore, with the proper adaptive algorithm, the filter can smoothly readjust itself under changing conditions to minimize the error. Examples of adaptive algorithms are least mean squares (LMS) and recursive least squares (RLS). <\/span><\/p>\n

    \"Figure<\/p>\n

    Figure 2. SE with adaptive noise filtering block diagram.<\/em><\/p>\n<\/div>\n

    Another example of multi-channel speech enhancement is <\/span>beamforming<\/span><\/i><\/a>, which uses a microphone array to cancel out signals coming from directions other than the preferred source. Multi-channel speech enhancement can lead to promising results, but it requires several microphones and is technically difficult. <\/span>
    \n<\/span> On the other hand, single-channel, or monaural speech enhancement, has a significant advantage because we don\u2019t need to set up extra microphone(s). The algorithm takes input from only one microphone, which is a noisy audio signal representing a mixture of speech and noise, in order to remove unwanted noise.\u00a0<\/span><\/p>\n

    The rest of the article is devoted to the monaural case.<\/span><\/p>\n

    One of the first results in the monaural case is the <\/span>spectral subtraction method<\/span><\/a>. There are various methods for this approach, but this is the idea behind the original method:\u00a0<\/span><\/p>\n