The following schema describes language identification in CUMULATIVE
mode.
[LangIDCumWav] 0 = a ← wav(MONO, input) 1 = f ← frontend(_, a) 2 = nf ← normalizer(_, f) 3 = lf ← lidfeature(_, nf) 4 = lid ← langid(CUMULATIVE, lf) 5 = output ← lidout(_, lid)
0
|
The wav module processes the mono audio data. |
1
|
The frontend module converts audio data into speech front-end frame data. |
2
|
The normalizer module normalizes frame data from 1 (f ). |
3
|
The lidfeature module converts normalized frame data (nf ) into language identification feature data. |
4
|
The langid module in the CUMULATIVE mode processes the language identification feature data (lf ) to identify the language. |
5
|
The language identification information (lid ) is written to the output file. |
The following schema is similar, but uses different language identification modes.
[LangIDBndWav] 0 = a ← wav(MONO, input) 1 = f ← frontend(_, a) 2 = nf ← normalizer(_, f) 3 = lf ← lidfeature(_, nf) 4 = lid ← langid(BOUNDARY, lf) 5 = output ← lidout(_, lid) 6 = output ← lbout(_, lb) DefaultResults=out
|