What Is High-Resolution Audio? Understand What “24-bit/192 kHz” Means

High-resolution audio is a term used to describe music files with 24-bit depth and a sampling rate above 44.1 kHz. In other words, tracks with a 24-bit/44.1 kHz resolution can already be considered high-resolution, although they are not classified as “Hi-Res” by the Japan Audio Society (JAS), which is responsible for defining the specifications required to obtain the “Hi-Res” label. According to the JAS, a file must be at least 24-bit/96 kHz.

Recommended article related to the topic:
• What Is Lossless Audio? How Music File Formats Work

What is high-resolution audio?

High-resolution audio is represented by files with a 24-bit/44.1 kHz resolution. The bitrate (data rate) of the music file must be at least 2116.8 kbps for it to be considered a high-resolution audio file.

What is Hi-Res audio?

For audio to be considered “Hi-Res,” the music must have a resolution of 24-bit/96 kHz, which is the specification defined by the Japan Audio Society (JAS). This means that the audio file must have a minimum bitrate of 4608 kbps to be classified as “Hi-Res.”

How to calculate the bitrate of a song?

To calculate the bitrate of a song, you only need to know its bit depth, expressed in bits (examples: 16-bit, 24-bit, 32-bit), and its sample rate, which commonly appears as 44.1 kHz, 48 kHz, 88.2 kHz, 96 kHz, 176.4 kHz, or 192 kHz. With this information, you can use the following formula:

Sample rate × bit depth × 2 channels (stereo) = Bitrate

Example:

192 kHz × 24 bits × 2 channels = 9216 kbps

Are there practical benefits to using high-resolution audio files?

According to Reiss, J. D., in his article “Meta-Analysis of High Resolution Audio,” published in 2016 in the Journal of the Audio Engineering Society, when standardized training is applied, it is possible to perceive a small difference between CD-resolution files (16-bit/44.1 kHz) and high-resolution files (24-bit/44.1 kHz or higher).

In the conclusion of his meta-analysis, Reiss states:

“Overall, there was a small but statistically significant ability to discriminate between standard quality audio (44.1 or 48 kHz, 16 bits) and high resolution audio (beyond standard quality). When participants were trained, the discrimination ability was much more significant.”

He adds that the duration of the test may be important in enhancing the ability to distinguish high-resolution audio from CD-quality files:

“Careful selection of stimuli, including their duration, may play an important role in the ability to discriminate between high resolution audio and standard resolution audio.”

Unfortunately, Joshua mentions that he could not cover many practical aspects of perceiving differences in high-resolution audio:

“Several important practical aspects of high resolution audio perception could neither be confirmed nor refuted. Most studies focused on sample rate, so the ability to discriminate higher bit depth, for example, 24 bits compared to 16 bits, remains an open question. None of the studies included in the meta-analysis used headphones, so questions about how headphone presentation affects perception also remain open. The meta-analysis also did not address issues related to specific audio system implementations, such as the choice of filtering applied, the specific high resolution audio format chosen, or the influence of various hardware components in the audio recording and playback chain (beyond evaluating possible biases that could be introduced by inappropriate choices).”

Mizumachi, M., in his article “Subjective Evaluation of High Resolution Audio Through Headphones,” presented in June 2016 at the 140th AES (Audio Engineering Society) Convention in Paris, France, described an experiment that used headphones to compare high-resolution audio files with music at near-CD resolution and MP3 tracks at 128 kbps and 320 kbps.

According to Mizumachi, the listening test included “36 individuals with different levels of experience in audio and music, covering a wide age range (20 to 70 years).” These participants were divided into two groups:

• Group A: composed of 11 audio enthusiasts and musicians who are familiar with live music performances and high-resolution audio files.
• Group B: composed of 25 individuals who typically listen to music in CD-quality formats and MP3, and who rarely attend live concerts.

Mitsunori stated:

“The appropriate selection of musical sources is important for perceptually detecting small differences between audio formats. The authors examined the suitability of several musical sources across various genres. A jazz piece is considered the most suitable for this type of perceptual discrimination, as each instrument can be heard independently and rapid temporal transitions frequently occur.”

The music track “Colors of Darkness” from the album “T-TOC DATA COLLECTION VOL.2” was used in a high-resolution file format at 24-bit/192 kHz. According to Mizumachi, participants listened to 120 seconds from the beginning of the track before comparing it with other formats and resolutions. He explains that the audio file was used as follows:

“The original HRA source was resampled and reduced in bit depth to a quasi-CD format of 48 kHz/16 bits. The data conversion was performed using the ‘resample’ function with a carefully designed FIR low-pass filter in MATLAB. Subsequently, the quasi-CD source was converted using the LAME MP3 encoder into two lossy MPEG formats: 320 kbps and 128 kbps, respectively. In the listening test, participants compared the same music in four different formats: PCM 192 kHz/24 bits (HRA), PCM 48 kHz/16 bits (hereafter abbreviated as CD), MP3 at 320 kbps (MP3-H), and MP3 at 128 kbps (MP3-L).”

According to Mitsunori, the equipment used by participants to evaluate the music was:

• Audio player: Pioneer N-50
• Headphone amplifier: Brüel & Kjær ZE0769-004
• Headphones: Sennheiser HD650

According to Mizumachi, the test procedure was carried out as follows:

“Paired comparisons were conducted among the four stimuli in the different audio formats. Participants listened to the music in two different formats with a 30-second interval in each paired comparison. In total, each participant was asked to select the stimulus with better quality from 12 stimulus pairs presented in random order. The correct response rate was defined as the proportion of times participants chose the richer format in each pair.”

The test results showed that Group A, which was more familiar with live music and high-resolution audio files, achieved a 77.3% correct rate when comparing CD-quality tracks with high-resolution formats. Group B, which listened to CD-quality music and MP3 files and rarely attended live concerts, achieved a correct rate of 46%.

Mitsunori also made a comparison using data from this study with another study he conducted using loudspeakers inside a car. The result can be seen in the graph below.

Mizumachi commented in his article on participants’ descriptions of high-resolution audio (HRA), stating:

“The perceptual advantages of HRA included spatial representation, richness of information, and natural reverberation. Participants’ comments on HRA with headphones did not mention presence and realism, although presence was one of the most important perceptual characteristics of HRA in previous loudspeaker experiments. It is assumed that headphone presentation is superior to loudspeaker presentation for discriminating details, as it can eliminate interference from background noise and room acoustics.”

The use of high-resolution audio in music streaming services

According to Melchior, Vicki R., in her article “High-Resolution Audio: A History and Perspective,” published in 2019 in the Journal of the Audio Engineering Society, lossless music streaming began to emerge because “consumer involvement stimulated the quality of online high-resolution releases” in several ways. Melchior cites the following factors as responsible for driving the adoption of high-resolution audio files:

• Demand for recording provenance: previously, many albums were released by upsampling files originally at CD resolution (16-bit/44.1 kHz).
• Requests for larger audio files: based solely on reports of incremental sonic improvements, without grounding in engineering and psychoacoustic principles.
• Use of the best available source when remastering older works.

Melchior states that “very high data rates may make sense for internal engineering paths,” meaning they are primarily justified during the recording, mixing, and mastering processes.

Even in 2019, Melchior observed an initial migration of streaming services toward high-resolution audio files. By 2025, it is clear that this movement has been widely adopted by most platforms, as the following options are now available for listening to high-resolution music:

• Amazon Music
• Apple Music
• Deezer
• Qobuz
• Spotify
• Tidal

However, despite the widespread adoption of high-resolution audio by streaming platforms, Melchior notes that discussions about whether higher-resolution music should sound more transparent have been ongoing for the past 30 years. In the end, she concludes that “the use of high-resolution formats does not guarantee the perception of transparency.”