Dave Plowman

Watched it last night. Quite a bit involved those wearing full PPE including those clear plastic face masks. And I could barely make out 1 word in 10. Of course as you get old your hearing (and or understanding) becomes poorer – but the elderly is their audience these days.

Nick Ware

My thoughts exactly. My hearing is in surprisingly good Nick (pun intended), but there was loads of last night’s episode that was far worse than it could/should have been. We both struggled to hear. Wife, Judi, being 20 years younger than me, and even she kept asking why it’s so bad.

Compare with almost any Sitcom, drama etc., from 30 years or more ago, and current-day standards are often pretty dire.

Newsroom presenters: invariably less clarity than live inserts from other countries far away. What’s the one thing you need in a news programme? To be able to hear clearly what’s being said.

I could go on. A topic rarely discussed on IPS groups for fear of criticising/offending sound colleagues. So it gets blamed on codecs, encoding, decoding countless times, finally decoded in your TV, so that’s where the blame ends up. The age old saying: “It’s OK leaving me”! But then, along comes another programme that sounds fine, which rather shoots any technical blame at my end down in flames.

Dave Plowman

One common answer is ‘it’s your TV’. But then you can listen to radio on FreeView too. And I have never had a problem understanding what is said on that.

I’d guess it’s the freelance thing. No one willing to stand up and say ‘this is unacceptable’ because trouble makers don’t get another contract.

Chris Wolfe

Yes, many things, but one that isn’t always mentioned is the coding quality. If we listen to something off air the sound quality can be acceptable – rarely better than that unless it is a “vintage” recording.

But listen to something else in the same series on catch-up and the intelligibility level goes right down for us. This has to be because the (re)coding of the audio suffers. In our case it is partly a very slow internet connection, which must need some heavy data compression/low bitrate to get a signal to us. But I’m not sure how many different quality levels service providers run to satisfy this sort of problem.

Mike Giles

May I suggest, in all seriousness, writing to the DG? I have done so recently and was taken seriously, but it would reinforce the argument if others also made comment on the unacceptable nature of much of the modern television sound output. Everything from soggy, unequalised personal mics on the news to mumbling artists in otherwise high profile drama.

I remember thinking, in respect of the landlord in Jamaica Inn, in particular, that had you been watching his performance through a shop window, with no sound audible, you would still have perceived that his delivery was pretty well unintelligible! That same drama had other examples of perfectly intelligible audio.

There is surely a sufficient body of well-informed former Beeb employees amongst adherents to this forum as to be capable of exerting influence where current practitioners fear to tread.

Dave Plowman

Due to the wall to wall sport, and being restricted in what I do due to a bad back, I’ve been watching more ‘old’ TV than usual. Let’s take an early Midsomer Murders. Pretty soggy pictures, but excellent sound. Later ones seemed to have reversed this, as far as the speech quality. And that is common throughout the industry. Of course some worse than others.

Being long retired, wonder who made the decision that poor quality speech was now acceptable? And it’s not entirely due using personal mics buried below unsuitable clothing, but the lack of careful EQ too. Especially given dubbing is a much cheaper process than once was the case.

Hugh Sheppard

Once upon a time… The BBC’s Chief Engineer/Director of Engineering sat – or probably stood – at the right hand of the Director-General.

Reith’s no. 2 was Captain Peter Eckersley from the inception of the BBC in 1923 until 1929, with Dr. Bill Dennay as the last D-E from 1987 – 1993. But what happened then? Step forward Baron Birt of Liverpool. D- G from 1992 – 2000 who, with the then Board, failed to straddle C. P. Snow’s recognition of the importance of both the humanistic and scientific cultures. We know the outcome; BBC Engineering disappeared into the wide blue yonder, taking the BBC’s once paramount operational standards with it.

OK; that’s just one former tech-ops point of view, but has anyone a better explanation of the real casualty behind the muffled speech of ‘Casualty’?

Dave Plowman

Simple, really. Personal mics are the cheap and convenient option – dispensing with skilled expensive labour. Very sad that so many on the Casualty production team (from office downwards) thought this acceptable. All to save the costs of a bit of easy post syncing.

Alan Taylor

Having recently chatted with a recordist currently working on prestige dramas and movies, it was interesting to compare notes.

He lamented that concealed personal microphones, each recorded individually was pretty well the only game in town. He would love to use overhead booms as the principal means of coverage, but it’s rarely possible when multiple camcorders are simultaneously recording every shot in close up and wide angle. There would also be challenges in maintaining consistent results when intercutting sequences using booms or radio mics. We used to face that same problem, but the other way around. Radio mics were only used when absolutely necessary. Virtually all coverage used to be done with boom mics or fixed spot mics. There were limited options for recording more than two tracks during a take either. These days, all decisions have to be made retrospectively and either sorted in the edit or glossed over. Digital recording for video and sound is now so ubiquitous and cheap that you can record every element individually all the time. On the face of it, that ought to be a good thing, but if that were the case, we should be celebrating progress, not joining the ranks of those moaning about unintelligible dialogue.

He seemed rather envious of the way we used to work. Two aspects really surprised him. One was that directors really would at least listen to what the sound supervisor says – even if they then chose to disregard it. The other was that we also did the post-production work too. He would have loved the opportunity to lead a project from pre-production, through shooting and then through the final dub, but it’s never going to happen.

The only time he expects feedback during post production would be if there was a serious problem. He never gets to learn about small things which work well, little things which caused minor headaches, or what could be done a little differently to improve the results another time. My belief is that you get best results by amalgamating a lot of tiny details, each incrementally only making a marginal difference, but collectively adding up to make a significant difference.

Alec Bray

Watching one news bulletin last week, the news anchor (who I did not recognise) introduced an item. If I caught one word out of ten, that’s possibly an overestimate. Could not work out what was being said. Cut to the insert – sound very clear, heard every word. (We than changed channels.)

I though it was just me, having a funny few moments.Yes, I have hearing aids now, usually they work fine. But some programmes are difficult to hear, even my wife turns up the volume.

Another issue is that all the streamers seem to have different sound levels, we are always having to adjust the volume when we change sources.

Nick Ware

If I could un-invent one thing – erase it forever – it would be the radio mic. OK, not entirely fair as they do have their uses.

A lot in my view, is down to the fact that sound recordists who make a living by supplying their own sound kit have always aimed to get as much kit on the job as possible. The “number of speaking parts in a scene = how many personals you need” approach made life too easy for Directors, Lighting and Actors, and ultimately, budgets, so there’s no going back.

Much of Casualty is pretty poor, but on the other hand, sound is consistently better on Eastenders, interiors in particular, for reasons that should be obvious.

Let’s next discuss how camerawork became so amateurish-looking, and the persistent obsession with non 16×9 cinematic aspect ratios!

Graeme Wall

There’s a whole generation of production staff who doesn’t even know there is an alternative to personals everywhere.

Don’t get me going on aspect ratios! After years of trying to reconcile 4:3 and 12:9 in the same shot so nobody would see a black edge they now all try to pretend they are Spielberg and use cinematic ratios on mundane productions. Then there is the over use of shallow depth of field: <fx: gets dragged off stage foaming at the mouth>

Dave Plowman

Thanks Alan – pretty well what I’d guessed. When I retired some 15 years ago we were already fitting radios even where not needed and sending isos to the recorder – but nothing like as many tracks as today. Because dubbing asked for it. But never more than 2 cameras then. Of course we all must have had a dodgy location/ situation where it simply wasn’t possible to get decent dialogue. And had ways round this. All of which cost time/money, and after all sound doesn’t much matter these days. Anyone can do it

Chris Wolfe

Sound levels get talked about by both techie and non-techie people, as do background music levels etc, but “intelligibility” hardly gets recognised as a term, let alone considered.

There are a lot of elements to intelligibility – simple signal to noise ratio, and clarity of consonants that divide up words, so making them recognisable, for instance. This latter can be affected by poor HF, which may be in the listener’s aged ear, or in muffled mics. Speed of delivery and mumbling also have a material affect on this.

Then there is ambience. The brain makes all sorts of distinctions about how it works out what a sound is, and much of this is based on ambience. The spectral response and level of reverb we expect when we have our ear close to someone for a whisper, or are 20 ft away hearing a shout is important. If the acoustic perspective from a personal mic is identical for both situations the brain struggles to unravel this.

These are problems that are not entirely in the hands of any one technician. Personals can be “helped” by recording local atmos and adding a measure of that, but this requires help from editors/dubbers sitting with a director who is only interested in the time it is all taking, and knows the script backwards, so never has to judge intelligibility.

And then there is delivery data compression and psycho-acoustic “shaping” that adds a further level of “issue” that the production team have no knowledge of or control over.

It is a complex problem, does need solving, but isn’t entirely simple to deal with.

Paul Thackray

I think the other thing that gets missed or not understood is the music to voice Balance is affected by the overall listening level.

The balance only works totally for the monitoring levels used when its mixed.

If you monitor loud (compared with the viewer) you will take the voice level down with respect to the music, so when the real view listens at normal levels, so will be buried.

It’s also common to use automation in post production to make sure the Loudness Specification is met. I have no idea what this does to the balance or if it varies by vendor l?

Alan Taylor

There are two points during the making of a programme where intelligibility can be seriously compromised. One is during the shoot and the other is during post production.

During the shoot, the way the actor delivers the dialogue is paramount. Nothing worthwhile can be changed by technology after that, other than re-voicing. Technical issues such as microphone choice, placement, EQ and background noise all become important parts of the equation. To the sound recordist / supervisor, one of the biggest challenges can be to persuade the director that there is even a problem with intelligibility. Their intimate familiarity with the dialogue allows them to ‘hear’ it perfectly clearly, while anybody listening for the first time wouldn’t be able to make sense of it.

A few times I’ve been able to convince a director that they are overly familiar with the dialogue. After a recording, I’ve asked the director to observe while we replay the take to a crew member who wasn’t particularly involved with the shoot, maybe somebody like a costume assistant, somebody who I imagine might represent a typical viewer. I simply ask them to listen on the headphones and repeat the dialogue out loud. It’s a very simple demonstration, but few directors are willing to “waste precious time” in that way. The other tactic was to do it with either the script supervisor or dramatist observing the replay. If they saw that the dialogue was not coming across clearly, they would support me when bringing it to the director’s attention.

The other time when dialogue gets compromised is in the dub. For obvious reasons, the director often wants to add powerful sound effects and exciting music. Again, over-familiarity with the dialogue is a big danger. At this stage in the production, they are even more familiar with the dialogue. In order to impress the director, sound effects such as explosions are likely to be auditioned really loud. Similarly music will be

listened to at high levels. In many cases these elements will be listened to in isolation, or without the dialogue being mixed in at a realistic level.

When it comes to the final mix, the director might well feel that the excitement which was previously there is now lacking. That’s when they want more of everything, with it all turned up to eleven. Sadly in the real world, there are limits to the maximum overall level possible for any given medium. On top of that, there can be issues where the sound effects or music are competing in the same part of the audio spectrum as the dialogue and therefore compromising intelligibility.

On those occasions when I’ve been able to liaise with the composer prior to them creating the music, I’ve tried to encourage them to feed the dialogue to one channel of their mixer and mix it to the monitoring speakers. The actual recording should be done from a mix consisting of everything except the dialogue.

I’ve worked with a few composers who have the output from the video ( dialogue ) fed to its own speaker. It’s much like feeding talkback to different speakers. You can discriminate between different sources spatially arranged around you, but that can also mean you’re able to pick out the dialogue rather better than if it is all mixed together. I’ve also tried to make composers aware of the acoustic spectrum. Really strident trumpets are a hallmark of a James Bond theme tune, but you would struggle to get much dialogue heard behind them. If dialogue and music are to co-exist, the theme needs to be carried with a little more subtlety, maybe with mellow instruments until there is a gap in the dialogue where the composer can really go for it.

There are no simple solutions and few technical magic bullets to improve it later. If the original recording is indistinct, it’s always going to be indistinct. If the original recording is nice and clean, it can still end up being obscured in the final mix. Obviously there are technical considerations to bear in mind, but by far the most important factors are the choices made by the actors, recordists, directors, composers and post-production people. Some of those judgements can be deeply compromised by people being too familiar with the script.

Nick Ware

All true, and I think there is another factor that compromises intelligibility – and that’s lip-reading. Intelligibility is as much about subconsciously watching facial expression, and lip movement in particular, as it is about sound alone.

My elder sister and brother-in-law, both in their late 80’s, repeatedly ask me why in God’s name do we have to suffer those absurdly wide newsreader and weather presenter shots that mean that you can’t lip read. To them it makes all the difference, and to some degree I think it does for all of us.

I would add to that, dingy lighting in dramas, and wobbly camera operation. I don’t know about you (anyone), but my head doesn’t shake around all over the place, and if I’m in real-life conversation, others around me don’t either. Because the image is contained in a static frame (often exaggerated by a black band top and bottom) the random wobble is much more objectionable than in a darkened much bigger cinema situation, if that’s what they think they are trying to aspire to.

By the way, it amuses me that subtitling often comes up with some interesting mis-spelling, because the AI can’t understand the audio either!

Alasdair Lawrence

The role of lip-reading is very important, and I recall that during Covid, I had some appointments at various NHS establishments, and often found it difficult to understand some staff who had the standard face mask on.

Coupled with different accents and speech patterns as well as unknown (to me) technical terms, it made life difficult especially as the system was, then and now, understaffed and over-worked.

Nick Ware

And, if it’s being dubbed in Dolby surround (music mainly to front left and right, dialogue to centre) it’s probably not being monitored in mono or stereo. The 5.1 surround is transmitted as such, the fold-down being done at the receiving end. How effectively the latter does that is anybody’s guess.

Here at home our main TV feeds two active pro monitors (2.0).

In my “man cave” which is mainly used for music editing, etc., I have two Mackie HR824 active monitors carefully matched with a Genelec sub. If it doesn’t sound good on that setup there’s something very wrong incoming, not here! Proof of that being that some programmes sound significantly better than others.

Last night, a several year-old Death in Paradise sounded fine, I suspect dubbed stereo, not 5.1

Hugh Snape

You’re dead right about recording the best possible dialogue quality during the shoot. Although it can be hugely improved with appropriate eq and compression during the dub that will always be something of a losing battle.

After I was made redundant from ITV I had an interview for a dubbing mixer job at a company associated with Sky. One of the questions asked was what I might do about material with poor intelligibility. Obviously I spoke about the various tools available which might help, and also the potential for post synching, but I also said I would be keen to speak to the production team to see how things might be improved at source; difficult to “make a silk purse” etc. This went down very badly, “surely the responsibility for improving audio lay strictly within the confines of the dubbing suite.” I suggested that it was the duty of everyone involved in the production to make things as good as they could be, but no ice was melted by my remarks.

Needless to say I didn’t get the job. I’ve always felt that that particular strand of the interview contributed in large part to my failure but that it also revealed a way of thinking about programme making which was not conducive to getting the best result.