Sound issue when capturing Hi8 through MiniDV camcorder

[Ad_astra]

Hi, I just captured a ton of Hi8 footage using the DV out function of a MiniDV camcorder. Sounded ok on TV, but when I used headphones to listen to a couple of AVIs, it turned out that all sound was coming from one channel. I tested the Hi8 source and it was stereo. When looking up properties of the captured AVIs it says it is AAC stereo 32bit.

What I discovered is that in the DV out menu, the MiniDV camcorder had source “1” or “2” selected, not stereo. The manual doesn’t have a lot on it, but it appears this has to do with mixing sound from two sources. When I subsequently captured with stereo selected, the sound was normal (both channels).

Hard to tell if any audio was lost or if everything was simply dumped on one side. Question is, if it still says Stereo in the properties, is there a way to separate it into two channels and do a batch job of that? Having to re-capture the whole thing would be a bummer.

Thanks!

[Alwyn]

Out of interest, what model is the camcorder?

I'm surprised the captured audio is AAC. How'd you capture the tapes?

[Ad_astra]

Alwyn,

The tapes were played on a Sony TR910 Hi8 camcorder connected via SVHS and two audio cables to a Sony TRV11 MiniDV camcorder. The latter was in a DV out mode and was outputting via FireWire to the computer with WinDV.

I just re-checked AVI properties. In fact, it doesn’t say AAC. It just indicates the bitrate (1024kbps), channels (2, stereo) and sample rate (32000khz). So, it looks like it is still stereo, it just happens to come out of one earpiece. Question is, how do I separate the audio into two channels.

[Alwyn]

Hmm, a tricky one. I think it would be best if you attached a sample DV file here so the audiophiles can examine it. The forum takes 500MB, but probably best to keep it down to around 100MB, which would be about 20 seconds worth. Pick an area with good sound.

[aramkolt]

If I had to guess, the issue probably is between the Hi8 camera and the MiniDV camera, most likely the input cable you are using to the TRV11 either not making good contact, or not having the correct pinout.

The RCA outputs you can hook to whatever from the TR910 to verify that both are outputting correctly and just disconnect one at a time and see if the speaker plugged into a TV switches.

[timtape]

It could have been a faulty audio cable, faulty camera audio socket or plug not pushed fully home. I think the trv11 has a headphone socket to check while digitizing. With many modern TV monitors it could be working on only one channel and it's not that obvious.

[Ad_astra]

The problem is not a malfunction but a setting in the MiniDV camcorder. There are three options - Stereo, ST1 and ST2. When I captured all of the hi8 tapes, it was set to either ST1 or ST2, but not stereo. When I later tested the capture with stereo selected, everything was normal.

Again, since the audio still shows up as 2-channel stereo in the properties of the AVI, I’m wondering if I can fix this somehow to make it come through both headphones/speakers rather than one. Would hate to have to recapture all of the tapes again just to get the sound right.

[Alwyn]

Attach a sample.

[timtape]

It's not just having the sound through both speakers. If it's a stereo source it must remain stereo. When capturing, the ability to hear the audio in true stereo downstream of the DV link is important. Many TV speakers are too close together to allow us to hear the problem while capturing. Metering also helps. A true stereo source will usually show up as constantly differing levels on a stereo meter.

[The_Doman]

The problem is not a malfunction but a setting in the MiniDV camcorder. There are three options - Stereo, ST1 and ST2. When I captured all of the hi8 tapes, it was set to either ST1 or ST2, but not stereo. When I later tested the capture with stereo selected, everything was normal.

About the audio being 32000 Hz:

Also, why does it capture audio at 32 kHz instead of 48 Khz?

Probably because you have (mis)configured your camera audio setup to use 32kHz audio mode?
Works the same with my D8 camera when capturing using analog-pass through.

Wikipedia: DV (video format)

Audio can be stored in either of two forms: 16-bit Linear PCM stereo at 48 kHz sampling rate (768 kbit/s per channel, 1.5 Mbit/s stereo), or four nonlinear 12-bit PCM channels at 32 kHz sampling rate (384 kbit/s per channel, 1.5 Mbit/s for four channels). In addition, the DV specification also supports 16-bit audio at 44.1 kHz (706 kbit/s per channel, 1.4 Mbit/s stereo), the same sampling rate used for CD audio.[11] In practice, the 48 kHz stereo mode is used almost exclusively.

Again, since the audio still shows up as 2-channel stereo in the properties of the AVI, I’m wondering if I can fix this somehow to make it come through both headphones/speakers rather than one. Would hate to have to recapture all of the tapes again just to get the sound right.

Possible with Virtualdub you can adjust/choose the channels?
From some older topic:

Lossless Quicktime to Regular Lossless AVI Conversion.


[Attachment 82829 - Click to enlarge]

[Ad_astra]

OK, I'm uploading a few seconds of what I'm talking about - audio in one ear which nevertheless shows up in properties as 2-channel stereo.

Thanks again for all your input.

Just to clarify, after playing with the MiniDV camcorder settings, now I know how to get the "correct" result (normal stereo which you can hear in both earphones/speakers). Problem is, I don't want to re-capture 20+ tapes which have been captured using the wrong audio setting. It's not just the copying, it's a lot of time spent with Scenalyzer. So, if there is a way to make it "normal" stereo using some kind of software, I'd rather do that, even if I had to process each file individually.

Appreciate your thoughts.

[Ad_astra]

Doman,

Just tried your suggestion - in VD, the only way I was able to get the sound in both channels was by going mono. Anyway, appreciate your idea.

[davexnet]

Seems to be technically a stereo file, but it looks as if the left channel has nothing in it.
You could make a two-channel mono file out of it

[Alwyn]

Yes, left channel flat-line...


[Attachment 88055 - Click to enlarge]

[The_Doman]

Problem is, I don't want to re-capture 20+ tapes which have been captured using the wrong audio setting. It's not just the copying, it's a lot of time spent with Scenalyzer. So, if there is a way to make it "normal" stereo using some kind of software, I'd rather do that, even if I had to process each file individually.

If the one / left track audio is good enough?
May be the source was mono too?
If your author/encoding software allows it you could just select the good channel and encode as mono or duplicate to stereo?
Other option is to extract the audio and process separately?
That will not take much time.
Then you can use/import the audio file when encoding your final video.
I have done that process many times in the past when making DVD's with TMPGEnc DVD Author from TV/VHS recordings.

[timtape]

Even some Video8 camcorders recorded and played back stereo. I suspect most or all Hi8 camcorders recorded and played stereo from the two mics built into the camcorder. Sometimes strong wind at an angle will hit one of the mics, totally wiping out the sound on that track but not so much on the other mic. It's good to always capture/digitize both mics and on separate tracks just as they were recorded, giving us maximum information and flexibility in editing.

In this situation of having digitized 20 cassettes it will be impossible to know how serious has been the loss of the left track unless we go back to the tapes, listen and compare.

Again, just as we monitor the picture as we digitize, so we monitor the audio, listening via proper stereo speakers or stereo headphones. That way, in the first seconds of the first cassette played, we should hear instantly audio problems such as a lost track, or both tracks accidentally joined together.

[The_Doman]

Again, just as we monitor the picture as we digitize, so we monitor the audio, listening via proper stereo speakers or stereo headphones. That way, in the first seconds of the first cassette played, we should hear instantly audio problems such as a lost track, or both tracks accidentally joined together.

If course this.
But also keeping an eye on audio level indicators during capture will show immediately if something is really wrong with with the audio settings/cabling.
So many times this has prevented mistakes for me during my many pass-through/capture sessions with Scenalyzer.

[rgr]

Alwyn,

The tapes were played on a Sony TR910 Hi8 camcorder connected via SVHS and two audio cables to a Sony TRV11 MiniDV camcorder. The latter was in a DV out mode and was outputting via FireWire to the computer with WinDV.

I just re-checked AVI properties. In fact, it doesn’t say AAC. It just indicates the bitrate (1024kbps), channels (2, stereo) and sample rate (32000khz). So, it looks like it is still stereo, it just happens to come out of one earpiece. Question is, how do I separate the audio into two channels.

I think you can do this in VDub using Advanced Filtering, but I haven't tested it.

https://gcdnb.pbrd.co/images/TVQJofSz3Yv0.png?o=1

But it's better to use an AviSynth script, at least to deinterlace your video.

And you don't actually need two audio channels. If you have one, it will play on both speakers anyway. And in the Audio conversion tab, just select "Left channel".

[JVRaines]

Problem is, I don't want to re-capture 20+ tapes which have been captured using the wrong audio setting. It's not just the copying, it's a lot of time spent with Scenalyzer. So, if there is a way to make it "normal" stereo using some kind of software, I'd rather do that, even if I had to process each file individually.

As a Windows CMD with ffmpeg.exe in your PATH:

for /f %F in (pathtomyvideofiles\*.avi) do ffmpeg -i "%F" -c:v copy -af "pan=stereo|FL=FR|FR=FR" "pathtomyoutputfiles\%~n1.avi"

[Alwyn]

@JVRaines, does pathtomyvideofiles need to be in quotes?

[johnmeyer]

I own the TRV-11 and have used it over the past 20+ years to transfer hundreds of tapes.

Capturing only one channel is almost certainly from incorrect cable connections.

Getting low bit-rate audio is the result of using the wrong TRV-11 setting. You want to set that to 16-bit. The default, for reasons I cannot understand, is 12-bit.

If you don't want to recapture, just use any NLE or any software of your choice to copy one channel to the other. You will then get audio from both speakers, although it will be mono. Depending on your original Hi8 camcorder, that may be all you have, so this result is exactly what you'd get if you were to recapture. Even if the original was stereo, the imaging that you get with two little microphones mounted less than 1 cm apart is minimal, so it won't matter. The low bitrate won't matter much either because audio quality with consumer camcorders was not great.

Thus, you should be able to proceed very quickly.

As for batch conversion of the audio into two tracks, you can do that with lots of programs, but I suspect that step will not be needed because I assume you are not going to leave your captures as DV AVI files and instead will eventually convert them to MP4, MKV, or something similar. Do the audio fix when you do the encoding, as part of that workflow.

[rgr]

Even if the original was stereo, the imaging that you get with two little microphones mounted less than 1 cm apart is minimal, so it won't matter.

Theoretically you are right, but manufacturers used a trick of additionally amplifying the "louder" microphone, which enhanced the stereo effect.

[timtape]

...Even if the original was stereo, the imaging that you get with two little microphones mounted less than 1 cm apart is minimal, so it won't matter...

Yes the recorded stereo effect will not be large, but that's not the only issue here. In the digitized copy the left channel was completely missing! All the audio came out of the right channel only. That is not subtle. But this wasnt noticed even after digitizing 20 cassettes. Assuming normal hearing it would probably be due to inadequate stereo monitoring. Perhaps effectively mono monitoring. Perhaps no audio monitoring.

The issue is compounded when if they do have some audio monitoring, some people only listen on the monitor screen's' own internal speakers, which are often rearwards facing and of course the speaker separation cant be wider than the monitor screen itself. The closer the stereo speakers are to each other, the closer our heads need to be to them. There is a 'triangle' we should be sitting in to fully hear the stereo effect, however large or small. Once we are set up like this, gross faults such as one channel completely missing will be obvious at the get go.

It's important that not only the video but also the audio be monitored properly while it's being digitized. We should hear what's actually going down to digits.

You say the loss of the left stereo track here "doesnt matter" but none of us has heard the original stereo sound on these 20 cassettes so none of us is in a position to be dogmatic here on what we havent heard, I suggest.

The OP might like to redigitize say one representative cassette with proper stereo sound and judge for themselves. They might also upload a stereo sample for forum members to form their own (now informed) opinion.

[johnmeyer]

It is true that I was making the assumption that the OP's tape was like 99% of the hundreds of 8mm, Hi8, VHS, SVHS, and Beta tapes I have transferred over the years, all of which contain right and left channels which have roughly the same volume level and same audio quality.

I'm just trying to help the OP get the job done. I have helped, in person, dozens of people transfer their own tapes and I have only seen a handful of those who were actually able to complete the job. Once you have more than a dozen tapes, each holding a couple of hours that needs to be transferred, most people bail out before they finish. The last thing I want to do is add more obstacles.

It sounds to me like this person has actually finished the capture. If he wants to recapture one tape to see if there is a massive difference when both audio channels are correctly transferred, that would certainly be a good idea. All I was saying is that I expect the difference between simply copying the existing channel compared to spending hours recapturing the entire tape, that "massive" will not be the correct adjective, and that "minor" or "indiscernible" will more likely be the result.

[timtape]

Fair enough but nobody suggested capturing both channels would likely represent a massive improvement. Rather it would be just capturing all of the recorded audio on the tapes. It's only complicated here because the error was apparently discovered after all the tapes had already been digitised.