I've just started a new video project with CIN and I'm using as source files an mp4 video at 50 fps whose audio track is at 44.1 Mhz, captured with OBS. At the same time as the video, I recorded an audio file with an external microphone in another program, which is also in 44.1 Mhz. It is set in CIN exactly to the audio track of the video.

I set the video format in CIN to 50 fps and the audio to 44.1 Mhz and it worked well for my first rendering tests.

But before going any further I'd like to know if you'd advise me to switch to 48 Mhz for the audio in CIN, knowing that I can easily convert my audio files to 48 Mhz and that in any case these are preliminary tests and that I'm likely to make other recordings? Or can I stay at 44.1 Mhz without any problem? (The video is intended for broadcast on Youtube)

Thank you in advance for your help.

Under Linux Ubuntu 20.04 with the Cin version in the repositories, I have a lag between the sound and the playhead. However, I have verified that my wav file is in 48 Khz and set similarly in the format parameters of Cin. Still in the settings my video is set to 59.9401 images / second, which is the original format of my video which is mp4. For the sound parameters, the problem is the same whether I am using pulse audio or Alsa with a Multiface 2 sound card, and when I go from 4096 to 2048 samples in the playback buffer, it doesn't change anything. Lagging the audio does not improve the situation either.

Do you have any idea where the problem might be?

Thank you in advance for your help.

Philippe

Maybe I missed something obvious but I can not figure out how to use the audio effect denoise fft.

I suppose it works similarly to how denoise works in audio tool like Audacity. You need to specify a time window in a track that contain noise and specifty how much "reduction" of that noise you want. Audacity has a three dials or so to control different parameters.

In cinerella-gg, the graphical tool shows one dial for "denoise power", has one option for the number of sample to use as reference. The dialog shows that the keyframe is sthe start of the reference. But how do we specify keyframe?  I assumed that the golden key that appear in the time line on the effect bar would indicate the beginning sample. But however I use that effect, the sound in exactly the same.

I was able to process the sound in Audacity, but it would make my workflow easier to be able to do it from cinelerra directly.

Bonus question, can we apply the same effect on different tracks?

Thanks a lot!

Andrea_Paz responds:

For compressors you can try watching videos from the creator of Cinelerra: Adam Williams.

Single Band Compressor
https://www.youtube.com/watch?v=CO_4h48w1BA

Multi Band compressor

https://www.youtube.com/watch?v=z0Enimwcm9k

You can also find other cinelerra plugins on his Youtube channel. Flanger, vibrato, chorus, tremomolo, reverb, synthesizer, etc:

https://www.youtube.com/user/heroineworshipper/videos

Michi responds:

thanks for the tip. I know these detailed videos, but the single band compressor does not deliver a good sound for me so far.
So far I have been working with Vegas Pro on Windows. Everything was quite simple there.
My preferred compressor setting for spoken language is:
Threshold - ca. 10 dB to 30 dB, depends on sound
Ratio - 3:1 or 4:1, sometimes more
Attack: ca. 7-10 ms
Release: ca. 20-30 ms
Compression ca.-5 dB bis -6 dB,
Output-Gain: as much as the material was compressed.

Have any of you had good experiences with the single band compressor?

fary54 chimes in:

I have tried many different compressors with many different settings.

Result: The only compressor I have a perfect result with is L_SC4.
You have to wet the shirt to find the settings that suit you.

For info, to incorporate sound effects, I use this setting:
RMS/Peak: 0,000
Attack Time: 21,014
Realease Time: $125,402
Threshold Level: 0,000
Ratio: 16,376
Knee Radius: 3,250
Make Up Gain: 10,286

Michi responds to this:

Hello to all who are looking for a good compressor. I have now found it after all: It is the f_acompressor.
The setting is adventurous because the numbers are partly very abstract values and do not always represent the value in dB. But with a little trial and error I now get very appealing results: See screenshot.

In contrast to the single band compressor, the quiet passages are no longer raised so much and do not fall back into totally quiet nothingness. I still have to try a lot, but with the f_acompressor, which is of course quite awkward to use, I will certainly have a good sound soon. This is enormously important for me on my plan to make all videos with Cinelerra in the future. Thank you and I am very happy that Cinelerra and this forum exist. Right now write me @fary54 and recommend the L_SC4 - I will try it right away and look forward to it.

https://drive.google.com/file/d/1t-Gl0Q9G1az4bFseRKmjNvw6-WQc1v4R/view?usp=sharing

User Michi provided generic settings for an Equalizer:

Frequenz Gain bandwidth
100 Hz - Inf. 12 Low Shelf
230 Hz -5,0 dB 2,0 Band
500 Hz -3,0 dB 2,0 Band
2.500 Hz +5,0 dB 3,0 Band
10.000 Hz -Inf. 12 High Shelf

Using the LV2_Calf_EQ_5_Band plugin I got a good improvement, maybe it became too "telephone".
Again Michi provided an example suitable for my voice, using CinGG's native plugin: EQ Graphic (see image1).

This leads to even better results.
As a complete beginner I wanted to ask some theoretical/practical questions about the use of Equalizers:
1- Do you examine the frequency spectrum of the entire audio file and intervene by trying to level out peaks and valleys?
2- If I lower the bass frequencies in the graphic EQ, at the same time the high frequencies are lowered, as if there is a low-pass filter. Is this normal? (see image2)
3- I've tried some de-essers but don't know how to make them work. Any advice?
4- Is it better to do sound editing inside CinGG or use external programs (Ardour, Audacity, etc.)?

EDIT: changed audio file.

https://blog.frame.io/2017/04/10/how-to-keep-your-sound-editor-happy/

It is very common when cutting to another scene to have the audio track cut over earlier (J-cut) or sometimes later (L-cut). Is there an easy way of doing this in Cin-gg?