Earth Notes: Metacast 1: Blue Yeti, Shy Eurasian Footwear (2020-04)

Updated 2024-04-04.
About podcasting - a new microphone for better sound.
251s "meta1" (captions) Uploaded . Downloads:
Transcript:

Hi, I'm Damon Hart-Davis, and this is an Earth Notes metacast for 10th April 2020.

Blue Yeti mic Blue Yeti back
I've been happy recording for this podcast with my MacBook Air's built-in mic. But as I accidentally discovered for my Sign-off for 2019 with my cheap mobile headset, I can do significantly worse. Given repeated prompts from some musician friends, I'm trying to up my game!

I hummed and hawed and then lashed out on a Logitech Blue Yeti USB Microphone which seemed widely recommended, if maybe more money than I necessarily wanted to spend (£88 including VAT and P&P) on a podcast with maybe less than one listener!

(Separately, to help deal with the inflated number of Skype and other calls provoked by lockdown, on the same day I bought a Jabra headset. Quite spendy for me.)

[00:52]

I'm quickly testing all four microphones available to me and that are candidates for recording future episodes. Though if I'm not at home I'm quite likely to revert to the zero-equipment zero-hassle MBA built-in mic for simplicity! We'll see.

For a test recording snippet I want something that exercises the gamut of my spoken language, ie English RP (Received Pronunciation). I was sorely tempted to use one of the diction exercises from the film "My Fair Lady" (based on George Bernard Shaw's Pygmalion) such as In Hertford, Hereford, and Hampshire, hurricanes hardly happen. But a phonetic pangram should be better.

Also, because people have complained about my sibilance and harshly suggested a "de-essing" filter, I can use a famous sound-check test phrase too.

[01:43]

My two-part test phrase is thus:

  • Testing: 1, 2, 3.
  • Are those shy Eurasian footwear, cowboy chaps, or jolly earthmoving headgear?

Listen to the sample text captured with each of the mics:

[02:47]

All are sampled at 44.1kHz, and captured and normalised in Audacity. All are stereo, except the Jabra.

I'd say that the cheapo [Leap] mobile headset mic is clearly the least good. I didn't even know that the headset had a mic when I bought it, so the mic was effectively a free bonus!

Next best is the new (mono) Jabra. I may not be doing it justice since the level seemed a bit low, making the normalisation work hard.

The MacBook Air's built-in (stereo) mic is clearly the best of the first three. It's done a good job so far for Skype, podcasts, and so on.

The Yeti sound has taken a bit of getting used to, and is less 'bright' to my ear than the Mac's, but is richer and warmer and better overall I think. (The sample is probably a bit harsh and even clipped.)

[03:38]

This brief episode is brought to you courtesy of the Blue Yeti. I think that it'll take a bit of effort to use well, eg getting gain and positioning right for my desk, and finding space to store it when not in use. But so far I'm happy with it.

Look in the show notes on this episode's page for the earlier samples, and a bonus sample or two when I get the opportunity to put the Yeti through its paces.

There's more on my "Earth Notes" Web site at Earth.Org.UK.

Show Notes

MacBook Air Leap mobile headset Jabra headset Blue Yeti

The mobile headset (£5!) in all its glory, a bit like listening through a hissy cardboard tube... But, it does work.

9s "test 1 mobile headset" Uploaded . Downloads:

The Jabra Evolve 40 Stereo headset (mono mic, nominal 100Hz to 10kHz).

8s "test 1 Jabra Evolve 40" Uploaded . Downloads:

The MacBook Air (6,2, ~2013) built-in mic, as all but one of the episodes to date have been recorded with.

10s "test 1 MacBook Air built in" Uploaded . Downloads:

Here is the first Blue Yeti sample, once the background hiss stopped... 48kHz sample rate (mic nominal 20Hz to 20kHz) first attempt in "cardioid" (forward-facing) mode, analogue gain minimum and digital gain cranked up, after it had "warmed up" a little and background noise levels dropped markedly

9s "test 1 Blue Yeti cardiod" Uploaded . Downloads:

The second Blue Yeti sample. 44.1kHz sample rate, analogue gain slightly up.

9s "test 1b Blue Yeti cardiod" Uploaded . Downloads:

16WW ambient garden sounds Blue Yeti sample, Good Friday lockdown afternoon. 44.1kHz sample rate, analogue gain at max, omnidirectional mode.

10s "20200410T1620Z Good Friday lockdown Blue Yeti omnidirectional" Uploaded . Downloads:

16WW ambient garden sounds Blue Yeti sample, Saturday lockdown morning. 44.1kHz sample rate, analogue gain at max, omnidirectional mode.

126s "20200411T0722Z 16WW 2:06 holiday Saturday 8am ambient" Uploaded . Downloads:

16WW ambient garden sounds Blue Yeti sample, Sunday lockdown just before 3am. 44.1kHz sample rate, analogue gain at max, omnidirectional mode. The normalisation is doing a lot of work here as the non-normalised signal is essentially inaudible.

75s "20200412T0245 16WW lockdown holiday Sunday 3am ambient omnidirectional max gain normalised" Uploaded . Downloads:

Sunday lockdown just before 3am, not normalised.

10s "20200412T0245 16WW lockdown holiday Sunday 3am ambient omnidirectional max gain" Uploaded . Downloads:

2021-12-26: Replacement: SC 165

202112 Jabra headset broken SC 165 USB headphones

My Jabra headset broke (one earpad and the mic boom snapped off) ~18 months in. The original Amazon Jabra supplier kindly sent me a direct substitute EPOS / Sennheiser SC 165 USB (MPC 508317, UPC 615104318038).

Nominal microphone response is 100Hz to 10kHz (cf headphones 20Hz to 20kHz).

I have been having what I take to be software issues with the audio volume from the mic dropping to inaudible levels for others on a call. Simply selecting a different audio source and then switching back immediately, with no fiddling around with cables etc, makes the audio come back.

The audio is more sibilant than the Jabra — here is a sample:

9s "test 20211226 EPOS SC 165 USB mono" Uploaded . Downloads:

Interestingly it was impossible to find anything about this SC 165 device on the Sennheiser site. (The EPOS search was bad also though.) Reporting the drop-out issue I received this in a response holding email:

Please note: Sennheiser Communications A/S, the joint venture between Demant A/S and Sennheiser electronic GmbH & Co. KG, has previously announced that it will evolve in new set-ups. Enterprise Solutions and Gaming headsets will evolve as a new independent company, EPOS, under the global audio and hearing technology group Demant. Going forward, EPOS will manufacture and sell the existing Sennheiser Communications portfolio for Enterprises and Gaming co-branded as EPOS | SENNHEISER products, while also introducing a new EPOS branded portfolio.

EPOS support called me on the morning of the 27th and suggested downloading their "EPOS Connect" app to update firmware on the headphones (from 44.05.18.18 to 44.05.31.31), and/or to use the 3.5mm jack directly.

EPOS support called me again shortly after I replied to them to see if there had been any improvement. I gently reminded the caller that it is still Christmas holidays here so I won't get a chance to test the fix(es) until Wednesday!

As of Thursday (30th) afternoon I have had a couple of workdays including various lengths of Skype and Teams calls using the SC 165. No signs of audio drop-out from the mic over USB — maybe it's fixed? (EPOS tried to call me again during one of those long Teams calls!)

I subsequently removed the huge (getting on for 1GB IIRC!) app.

Feedback

Tomasz said on Twitter on the 11th: Hey @EarthOrgUK, I'm sure you have more than one listener of your podcast! The quality improvement with the Yeti is fantastic, please don't use the built-in mic in your MacBook Air ever again ;)

Sources/Links

  • Logitech Blue Yeti USB Microphone.
  • Jabra Evolve 40.
  • SC 165 USB.
  • Diction exercises.
  • For the blue yeti how do I reduce background noise?: Gain on the actual microphone (the knob) all the way down to basically 3%. ... Noise Suppression ... Noise Gate ... The blue yeti picks up everything and you can't stop it fully. ... Digital gain will be your friend with this mic and you need to reduce your physical gain as low as you can honestly with this mic.
  • Blue Yeti Setup: even though I'm in Cardioid Mode, I still pick up loads of noise ... Set your mic 6-8" from your mouth. You'll want a pop filter 2" from the mic. Open up Audacity (free audio editing program), start a new recording, and talk at a normal level. Adjust the gain (mic volume) until you average about -12 dB at a normal talking level. ... Gain very low, mic close to the mouth.
  • Updated De-Clicker and new De-esser for speech (2014).
  • Is there a text that covers the entire English phonetic range?: English phonetic pangrams: That quick beige fox jumped in the air over each thin dog. Look out, I shout, for he's foiled you again, creating chaos and Are those shy Eurasian footwear, cowboy chaps, or jolly earthmoving headgear?
  • Audacity Tutorial: 17 Essential Podcast Recording & Editing Tips including de-essing with "Auto Duck".
  • Audio Quality Information & Standards for BBC Radio: Linear PCM (WAV or FLAC), 16 bit, 48 ksps, max true peak -1 dBTP, integrated loudness -23 LUFS, nominal 20Hz to 20kHz. Emission online profile 2a for MP3 is 128kbps stereo (not joint).
  • Podcasts: Mono Stereo or Joint Stereo?: Note that there is no downside to using Joint Stereo encoding for mono content — it will be encoded just as efficiently as if you'd forced everything to mono in your encoding settings. And the benefit is that any occasional stereo content's "stereoness" will be preserved, which is great for opening music, etc. I ... put everyone at slightly different points on a stereo soundstage ... cf The default recommended starting point is exporting the file as mono, 64 kbps MP3, CBR, 44.1 kHz, 16 bit, with a target loudness of -19 LUFS encoded with the Fraunhofer CODEC. If you need stereo for a drama, music podcast or walking tour for example, then 128 kbps (that's 64 kbps per channel), joint stereo and -16 LUFS.
  • Audacity MP3 Export Options: The choice of mono/stereo depends on the type of podcast. For voice only podcasts, mono is usually preferred ... 64 kbps CBR mono can give reasonable quality for voice if you are wanting to minimise the file size. 96 kbps CBR mono can give excellent quality for voice. 128 kbps CBR stereo can give reasonable quality where stereo is required. Also: Joint Stereo: Enabled by default, this button lets the encoder switch from frame to frame between "Stereo" (which just encodes the left and right channels independently) and Mid/Side stereo. ... This allows recordings with relatively little stereo separation to be encoded more efficiently ...
  • Why and How Your Podcast Needs Loudness Normalization.
  • Pics of all four mics.
  • Convert simple WebVTT body to HTML5 with an awk script.