These sound recordings compare cell phone headsets under similar conditions. As I use my cell phone primarily in my car, the tests include various driving conditions.
Impetus: As the TDMA network I'm currently on is being shut down, I'm having to upgrade to GSM. Not a bad thing, or so I thought. The tests include my old system: a TDMA(!) Ericsson R300LX, which is a cheap phone AT&T gave me for free when they phased out the technology of my previous phone.
The phone paired to the newer Bluetooth headsets is a Sony Ericsson S500i GSM phone. This is a slider phone, significant when used without a headset as the mic extends nearly to the mouth, and it affords being held so that the hand shields the mic from wind and some background noise. The network is AT&T GSM in California.
The moving tests were in my Mitsubishi 3000GT Spyder. With 320hp, there's no serious effort from the engine at these speeds, so minimal engine noise. With Yokohama dB S2 (ultra-quiet) tires, there's minimal road tire noise. I conducted the 35MPH tests on a somewhat rough road that hadn't been repaved in awhile, and the 60MPH tests on a concrete freeway in reasonable shape. This is a hard-top convertible, so sound with the top and windows up is comparable to that in a sedan (though it's a sports car, not a luxury quiet vehicle).
When inside (the house) and at 35 MPH with the top up, I spoke at normal volumes for the recordings, trying to be consistent. For the noisier environments, I spoke at only slightly elevated volume. Shouting might improve the signal:noise ratio.
My old phone was paired with a Plantronics boom, dual-ported, wired (non-Bluetooth) microphone. I can't find the number on it. I bought this because of good experience I'd had with my Plantronics aviation headset when flying small planes, which present a significant noise challenge. The boom comes down nearly to the edge of the speaker's mouth, which helps greatly in noise cancellation: the headset can compare the "signal+noise" from the port on the mouth side to the "noise" from the port on the outer side, with the difference being the speaker's voice. Without this information, the other headsets I tested are at a significant disadvantage, and performed significantly worse. So much for "upgrading". This headset does not have a wind sock, and suffered wind noise for it.
| Cardo Scala 700 | Logitech Mobile Traveller | Plantronics (with boom) |
no headset (S500i) |
|
| inside |
MP3 - Good accurate; some hiss |
MP3 - OK moderate hiss (not quite a hissy fit) |
MP3 - Great accurate; slight buzzing (from phone) |
MP3 - Good nasal tone; slight hiss |
| 35 MPH closed |
MP3 - Passable significant road noise |
MP3 - OK moderate road noise |
MP3 - Good some road noise when not speaking |
|
| 35 MPH open |
MP3 - Bad road and wind noise |
MP3 - OK road noise |
MP3 - Passable wind noise |
|
| 60 MPH open (windows up/down) |
MP3 - Unusable/Unusable significant wind and road noise |
MP3 - Bad/Unusable significant road and some wind noise |
MP3 - Bad/Unusable severe wind noise; voice relatively loud |
MP3 - OK/Passable nasal tone; some road noise |
My conclusion based on this limited sample is that no headset will perform well in a moving convertible with the top down without both wind and noise cancellation assist. Wind socks (some type of foam around the mic) seem common on outdoor, pro microphones and headsets, so I imagine that will be necessary. The Logitech Mobile Traveller has a hard foam which Logitech pitches as high tech. Based on the "35MPH open" sample, I believe it helps (as the Cardo Scala 700 otherwise beats out the Logitech). The Plantronics definitely suffers from no wind sock.
Noise cancellation seems best done with multiple signals to compare. The Plantronics I have has dual ports, as do most of the headsets which claim noise cancellation, which gives two signals to compare. Some headsets process that information electronically (typically DSP), which drains power, and others do it mechanically/acoustically.
A headset I'm very interested in is theBoom. It impressively deals with the noise inside a BlackHawk helicopter. It has a dual-ported mic on the end of a long boom reaching to the edge of the mouth, so has good information for differencing. (I say "long" boom to differentiate it from the ones that are a couple of inches long on so-called "boom" Bluetooth headsets.) Currently, theBoom isn't available as a Bluetooth version, but a comment on their Web site from a year ago indicated they were within a year of introducing one. theBoom uses only acoustic noise cancellation. It does have a wind sock.
One other headset I considered was the Jawbone. It has a third input from a sensor resting on the speaker's jaw, and uses that in differencing. It does not have a boom extending to the speaker's mouth, so doesn't have direct air-displacement differencing information as close to the source as a boom mic does. It's a cool looking designer accessory. After listening to all their demos, I was concerned about ensuring good contact with the jaw bone at all times (several users complained of difficulty with this on forum posts), but more about the voice being distorted in this artificial, metallic-sounding way that I noticed at times even in their recordings and read complaints of on forum posts.
The winner is: when not on the freeway, my old Plantronics boom headset; when on the freeway, no headset and windows up. I'll look for a connection adapter for the Plantronics that plugs into the S500i. Though Bluetooth was a requirement when looking for a phone, I may well end up not using it much for now.
I'm most hopeful about Bluetooth versions of theBoom and Plantronics with a boom and wind sock. An open convertible at 60 MPH is an extremely challenging environment for a consumer headset!