On 11/17/2025 10:01 AM, Erdinc Kaya via ffmpeg-devel wrote:
> - Correctly scale bitrate by bits_per_coded_sample instead of assuming 8 
> bits per byte.

When is a byte not 8 bits?

> - Prevents wrong bitrate reports for 16-bit PCM WAV files.

Do you have a sample that gets the wrong bitrate? I generated one just 
now, a 1 second 44100Hz 16bit stereo pcm wav file, and it gave me the 
correct bitrate of 1411 kb/s.

This function reads bitrate from a coded value in the container, 
BytePerSec, which is converted to bits. So not sure why you think it 
needs to be altered.

> 
> Signed-off-by: Erdinc Kaya <erdinc@veed.io>
> ---
>   libavformat/riffdec.c | 4 ++++
>   1 file changed, 4 insertions(+)
> 
> diff --git a/libavformat/riffdec.c b/libavformat/riffdec.c
> index 30835d5f36..d63edc6377 100644
> --- a/libavformat/riffdec.c
> +++ b/libavformat/riffdec.c
> @@ -130,6 +130,10 @@ int ff_get_wav_header(AVFormatContext *s, 
> AVIOContext *pb,
>               par->bits_per_coded_sample = avio_rb16(pb);
>           }
>       }
> +
> +    if (par->bits_per_coded_sample)
> +        bitrate *= 8.0 / par->bits_per_coded_sample;

bits_per_coded_sample is going to be 16 for a 16 bit pcm sample, 
clearly, so this here will halve the value in bitrate.

> +
>       if (id == 0xFFFE) {
>           par->codec_tag = 0;
>       } else {