Effects to improve the audio

I’m writing the effects i use in my audious and wan’t to get more professional advices about what should i do.
I start with noise reduction 2 times,then need to tell that ,my mic is quiet, so i start with normalization -6db add compression eq and the normalization -3db.Also started using talkspeech plugin with noise reduction only as an alternative.Want to hear should i use different order and should i use limiters?

The alarm bells are ringing. Whenever anyone says “noise reduction” as the first effect that comes to mind, it usually indicates that the original recording has way too much noise to salvage a “good” recording. You’ve taken that a step further saying “2 times” :astonished:

We should go back a few steps - What are you trying to record, what equipment are you using, and describe your method.
A short audio sample (just a few seconds), in WAV format,would help. Include whatever you are recording, and a couple of seconds of lead-in silence. The recording should be totally unprocessed (no effects - just a raw recording). See here for how to post an audio sample: https://forum.audacityteam.org/t/how-to-post-an-audio-sample/29851/1

In most cases just 3 or 4 seconds of audio in WAV format is sufficient.

I’m not sure about that. One or two words isn’t enough to tell when someone is doing something wrong.


If you do post in stereo, only ten seconds will fit.


Also started using talkspeech plugin

From where?

As above, it rings alarm bells when somebody arrives on the forum with a long list of effects and filters and never mentions the goal.


Thank you everybody for the attention,now answering the questions.
For recording using mic shure c 608.Want to record a review on a video series.Attaching the link with raw audio file
All the effects that i’m using learned from tutorials on youtube.Talkspeech plugin that i’ve mentioned is from nyquist

That’s way too quiet : the voice is only ~16dB louder than the noise floor.
You need to look at your hardware (gain) settings to boost the voice volume.

In the meantime you need to normalize

Normalize suggested settings.png
Then apply this code in Nyquist Prompt to get rid of the whine

(setf *track* (notch2 *track* 1499 30))
(setf *track* (notch2 *track* 3600 30))
(setf *track* (notch2 *track* 4096 30))
(setf *track* (notch2 *track* 9600 30))
(highpass8 *track*  20)

what the de-whine code looks like in Nyquist Prompt.png

I heard from tutorials that input volume should be 85-90% from max level when you record and shouldn’t use input gain in windows,something incorrect?

That’s cutting things a bit fine (very little “overhead”). It is essential that the recording level never exceeds 0 dB (100% full track height) as this causes irreparable damage. Unlike analogue recording, digital audio is totally unforgiving when it comes to overload. On the other hand, because the digital noise floor is extremely low (even for 16-bit audio), digital recording levels may be much lower than analogue recording levels without any loss of sound quality. We generally recommend that you aim for a recording level of around -6 dB, which is 50% of the track height.

Ideally, the recording level should be set by the hardware before it reaches the A/D converter, rather than digital scaling by Windows. In practice, due to the audio architecture in Windows, it can be very difficult to determine what Windows settings will achieve “unity gain”.

Some things that should certainly be checked are:

  1. Turn off “boost” in the recording settings of the Windows Sound Control Panel (if present)
  2. Turn off “Windows sound enhancements” (see: https://manual.audacityteam.org/man/faq_recording_troubleshooting.html#enhancements)

When using external (USB) recording devices, there is often some kind of LED metering built into the device, which indicates whether the level is correct in the hardware. For example, when using a microphone with a USB pre-amp, for voice recording, the usual set-up procedure is to speak very loudly into the mic, and adjust the recording level on the USB pre-amp so that the pre-amp’s “clipping” indicator just starts to illuminate, and then back off the recording level a little so that the clipping indicator stays off.

Once the hardware is set up correctly, the Windows recording level can be set to give a peak of around -6 dB in Audacity.

i used de-whine effect as you sudjested,should i use it in all of my audio/

The de-whine notches in that code are specific to your set-up, (e.g. probably won’t work on another computer).
The de-whine code only makes a subtle improvement …