Log in

View Full Version : The presence of silence in a audio file


Dave1024
26th August 2012, 18:50
Hi,

I had a audio file with one audio track and 2 channel.

My doubt,'

If the channel 1 had silence @ (sample[chn_index][sample_index] == 0), then channel 2 had the same 'sample_index' for silence?

where,
chn_index is channel count
[chn_index = 0 for channel1 and 1 for channel2] and
sample_index is sample count.

Thanks
Dave

tebasuna51
26th August 2012, 21:53
I don't know audio files with this structure.

PCM uncompressed audio files (.pcm, .wav, .w64, .aiff, ...) have: sample[sample_index][chn_index]
(all the channels of first sample, all the channels of second sample, ...)

Here "channel 2 had the same 'sample_index'" than channel 1 always, not only for silence values.

Dave1024
29th August 2012, 18:31
Yes I agree,

Will you clarify,
The duration(start and end) of silence(mute) or tone sound is similar in all the channels.? For eg: if multiple language audio is encoded in different channels of same track, start and end will similar ?

Thanks
Dave

tebasuna51
29th August 2012, 21:15
The duration(start and end) of silence(mute) or tone sound is similar in all the channels.?
Nope, each channel can be different.

if multiple language audio is encoded in different channels of same track
I never see that. Each language must be in a different track.

mbcd
30th August 2012, 07:15
I never see that. Each language must be in a different track.

Not in Germany, there was a time where some little TV-Station broadcast 2 languages in one Audiotrack.

Left channel was german, right channel was english, so they used a single Stereosignal to broadcast two languages as monosignal.

Silly silly ...:devil:

pandy
4th September 2012, 12:26
Why Dual Channel audio is silly?

Most older STB decoders have limited PID filter (ie part TS processor that extract PID from TS) use dual channel give opportunity to create 2 language with only one PID - for multilingual services in Europe there can be easily 12 - 20 languages active - dual channel allow use 6 - 10 PID instead 12 - 20 - (STM chips have limited amount hardware filtered PID to 16 and you need to have PID for video and various system services and data). Currently broadcasters shifting to single channel however dual channel is quite nice method to deal with some limitations (HW and SW).

pandy
4th September 2012, 12:27
I never see that. Each language must be in a different track.

nope
Dual channels service + ISO 639 descriptor + language descriptors

tebasuna51
4th September 2012, 14:46
nope
Dual channels service + ISO 639 descriptor + language descriptors
Then, just I never see that.

pandy
5th September 2012, 17:42
Then, just I never see that.


Quite common, currently broadcasters shifting from dual channel to single channel (mono) configuration.

Dave1024
8th September 2012, 19:31
Hi,

I am looking for measuring the loudness of audio stream based on the recommendation in
http://www.itu.int/dms_pubrec/itu-r/rec/bs/R-REC-BS.1770-2-201103-I!!PDF-E.pdf
My thought is, for the measurement i need to avoid the silence, mute and tone for getting a more accurate value to the channels present in the audio stream.
Is my thoughts about the removal of silence,mute and tone from loudness measurement from audio stream, will it give a accurate measurement or not?


Thanks
Dave