90% of all BangBros porn is AAC46. I think I have never encountered a "BangBus" video with more than AAC46.

Why? Most humans (and we exclude all persons below 18) cannot hear the difference between recordings with AAC46 or AAC96 when it comes to speech. That means: No background noise/music, no nothing ... just speech (or moaning)

AAC96 is enough for background music and videos like "Star Wars: Clone Wars". I never tested it, but for high quality music AAC96 might not be enough. IMHO AAC64 equals MP3-128 a standard for music between 1996 and 2002.

Hach! Star Wars Clone Wars ...



A major difference with AAC48 is music and sounds like an AC.

Facts

38:36 (38,5) minutes with AAC64 vs AAC96
AAC64
Duration : 38 min 36 s
Stream size : 17.8 MiB (1%)

AAC96
Duration : 38 min 36 s
Stream size : 26.7 MiB (1%)
... less than 10MB in 38,5 minutes.


AAC48 vs AAC64

AAC48

Audio
Format : AAC
Duration : 22 min 44 s
Bit rate mode : Constant
Bit rate : 48.0 kb/s
Maximum bit rate : 48.7 kb/s
Stream size : 7.91 MiB (1%)


AAC64

Audio
Format : AAC
Duration : 22 min 44 s
Bit rate mode : Constant
Bit rate : 66.2 kb/s
Maximum bit rate : 64.8 kb/s
Sampling rate : 48.0 kHz
Stream size : 10.5 MiB (1%)
... less than 3MB in almost 23 minutes.

I downloaded vids from Noodlemagazine with 240p, but the sound was AAC128 or even AAC312 (rare).

For example:
https://www.avsubtitles.com/movie136/scooby-doo-a-xxx-parody-2011

You can download it with 240p and AAC126 - which is very high just for the speech. The movie has a lot of sound effects and music of course, but I bet you would not heare a difference in AAC96 (or even AAC64).

DELETED by Admin Freejack: No double posts! Please do not post the same (identical) stuff twice. Please use hyperlinks instead. Thank you!

ORG-Post:
https://www.avsubtitles.com/bbs_posts.php?bbs_topic_id=4315#content_post_4378

Audio bitrate
In the case of audio files, the transfer rate also influences the quality of the reproduced sound. The more bits per second that can be reproduced, the higher the sound quality and the more detail we can perceive.

It is measured in kilobits per second (kbps) and directly affects the quality of the sound reproduced.

Thevideo resolution (1080p, 720p, etc.) only affects the visual quality and does not directly determine the parameters of the underlying audio file. Audio is a separate data stream within the container file (such as MP4 or MOV), and its parameters are what actually matter for speech recognition
Important audio parameters for speech recognition
For effective speech recognition, the following audio parameters are crucial:
Sampling Rate: This is the number of audio samples taken per second. A higher sampling rate (for example, 44.1 kHz or 48 kHz ) captures a wider frequency range and offers better audio quality, which is beneficial for speech recognition.
Bit Depth: Determines the dynamic range and accuracy of each sample. A greater bit depth (e.g., 16 bits or 24 bits ) results in a cleaner sound with less background noise, improving vocal clarity.
Bitrate: The amount of data used per second for audio. A higher bitrate (for example, 128 kbps or higher for MP3, or lossless formats like WAV) preserves more detail from the original audio and avoids aggressive compression that can degrade the voice signal.
Audio codec: The algorithm used to compress and decompress audio. Codecs like AAC-LC or MP3 are common in videos, but for speech recognition, it's preferable to use codecs that minimize quality loss or uncompressed formats like WAV, if possible.
Volume (Gain Level): An optimal audio level, which avoids clipping (peaks) and is loud enough for the voice to be clear over background noise, is essential. It is suggested that the voice gain be kept between -3 and -6 decibels.
Signal-to-noise ratio (SNR): This is a critical factor. The clearer the voice compared to the background noise, the better the speech recognition software will perform. Using a good microphone and a quiet recording environment is more important than video resolution.
Is the audio the same for all resolutions?
Yes, generally the audio file parameters are the same for all different video resolution options (1080p, 720p, etc.) if you are viewing the same source file or different encodings of the same master file on a platform (such as YouTube or Vimeo).
Video resolution refers solely to the number of pixels in the image. When a video is uploaded to a platform, the audio is processed and encoded separately, and that same audio track is typically used for all available video qualities.
In summary, focus on the quality of the original audio recording (microphone, environment, levels) and its technical parameters (sampling frequency, bit depth, bitrate) for good speech recognition, as the visual resolution of the video has no direct impact on the sound.