What is the difference between narrowband and wideband, closetalk and distant mic?

Maybe you have coped with terms like narrowband, wideband, closetalk, distant microphone, microphone array, and farfield in past. So let me explain it a bit.

All of these terms are about the "technology" you are using for recording the speech and its relative placement against the speaker.

Why do we need to bother with this? Actually the problem is that a speech recognizer (or a generic speech technology) is trained on data recorded under specific condition (telephone conversations for example). So this recognizer will recognize telephone conversations well, but will perform poor on lecture recordings recorded with a camera microphone in a room with strong echo.

As the research in the speech technologies field goes on, the recognizers are more and more robust. So this problem will shrink in future. But it still holds - if your data matches the data on which the recognizer was trained, you get the best possible accuracy. There is not acoustic mismatch.

There are, let's say, three variables:

• Quality of the recorded audio - sampling frequency
• Distance between speaker mouth and the microphone
• Number of microphones - microphone array

Narrowband vs. Wideband

Sampling frequency is one factor which can decrease the quality of recorded audio and the final accuracy. There are two settings - 8kHz and 16kHz (and more). If the data is recorded in the 8 kHz, it is so called Narrowband data. This settings are used in telephony. So if you work with telephone recordings, your data is in 8kHz due to the telephony technology limitations. Recording the telephone call in 16kHz or more does not make sense (and bring no improvement).

 An example of narrowband - 8kHz data

New Czech recognizer online

We are happy to announce that our new Czech recognizer is online at spokendata.com. The novelty lies in the support of 16kHz audio data (so called wideband) and also in robustness for distant microphones.
In comparison to the previous Czech recognizer, which was aimed at telephone data (8kHz - so called narrowband), the new one achieves better accuracy for:

• Speech recorded with higher quality - 16kHz or more. Unless you capture a telephone call, you usually record the data up to 44kHz.
• Distant microphone recordings. A telephone call recording is usually labeled as a close-talk because the speaker's mouth is close to the microphone. The distant microphone is the opposite case. Here the speaker's mouth is far from the microphone. As an example it can be a voice recorder or a mobile phone lying on the table while recording a dialog.

Fell free to test the recognizer by uploading your data and enjoy the transcript.