Answer the question
In order to leave comments, you need to log in
Asterisk - how to improve the sound quality of a file?
1. We record the sound in high quality WAV format
2. We convert the file on the server using FFmpeg to wav 8000 Hz 128 kbps
3. In the autoinformer, the file sounds quiet, if music is superimposed, then it is even less legible
4. If you call through the same SIP channel with phone to phone - conversation is loud and clear
[zebra]
type=peer
host=sip.zebra.ru
usereqphone=yes
context=outbound
dtmfmode=rfc2833
compensate=yes
relaxdtmf=yes
disallow=all
allow=alaw
allow=ulaw
allow=g729
Answer the question
In order to leave comments, you need to log in
I didn't really understand exactly how you listened to the autoinformer. It's also a good idea to listen to the recordings of the informant and the conversation.
So far, I can only make this assumption: all telephone systems amplify the signal non-linearly, namely, they compress it: the lower the signal level, the greater its amplification, as a result, the volume is “trimmed” and the volume range is compressed; hence the name of the process.
(More precisely, we are talking about the dynamics of the sound; the difference between the quietest and the loudest is called the dynamic range. A compressor is a kind of dynamics processor, it compresses the dynamic range.)
And you didn’t process the record for the autoinformer in this way. Try processing with a compressor - it is available in any sample editor, including the free Audacity. You need to compress before overlaying music, the exact parameters depend on your signal, but start choosing them with the following: all time parameters (inertia) to a minimum, the ratio is 1:6 (for 6 dB difference in the input signal there will be 1 dB difference in the output).
For
Didn't find what you were looking for?
Ask your questionAsk a Question
731 491 924 answers to any question