Timeline for Trouble understanding the speech of overseas colleagues
Current License: CC BY-SA 4.0
8 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Mar 29, 2019 at 6:04 | comment | added | virolino | @MSalters: Nobody ever transmits a 20kHz range of frequencies ever!! I am not talking about the spectrum of what you can hear, I talk aboutwhat is transmitted. CD's, DVD's and other audio recordings are also out of the discussion, as they cannot be really considered "communication" - not in the sense used in this question. I have an MSc in telecommunication, I know what I am talking about. AND number of bits has nothing to do with yelling. | |
| Mar 28, 2019 at 15:17 | comment | added | MSalters | @virolino: Nonsense. Given that it's voice, the frequency is already band-limited to 20kHz at best. "Bits per sample" can be converted to dB at 6 dB per bit; 16 bits gives 96 dB. You have to yell really hard to exceed that. That gives us L16 encoding at 44kHz, which is RTP/AVP standard payload #11. | |
| Mar 27, 2019 at 23:29 | comment | added | Harper - Reinstate Monica | Yeah, this doesn't eliminate the analog-digital-analog loop, just uses the phone company's choice of codec instead of yours. | |
| Mar 27, 2019 at 23:29 | comment | added | David Jones | However, in my case, it DID make a difference. The colleague was oriental, with a heavy accent, that somehow confused the linear predictive codec, resulting in gibberish. Over a clear (by North American standards) line, the accent was heavy, but I could easily make out what he was saying. | |
| Mar 27, 2019 at 13:27 | comment | added | Lightness Races in Orbit | I find landlines to be among the worst forms of "telephone" communication in terms of fidelity, particularly long-distance. An odd choice. | |
| Mar 27, 2019 at 11:37 | comment | added | virolino | ALL voice communication uses compression, one way or another. Starting from limiting the audio frequency spectrum, to choosing a sample frequency as low as possible, to choosing the data bits per sample, to very complex digital algorithms. Nowadays, most communication is digital, so "does not compress the audio in any way" is a technical impossibility. Moreover, the question is about the way the colleagues from India speak English (with specific accent), and it is not about the quality of the communication line. | |
| Mar 26, 2019 at 22:00 | review | First posts | |||
| Mar 27, 2019 at 11:28 | |||||
| Mar 26, 2019 at 21:38 | history | answered | David Jones | CC BY-SA 4.0 |