There's two variables with the same name:
1) in struct PHONEME_DATA, used by *phdata, *phdata_next and *phdata_prev
2) an unnecessary one that's only being memset() twice but never used
   for anything

f_log is only used in compiledict.c.

See commit 65847bd251. Bool recursing is not used for
anything anymore. Fixes a (compiler warning.

It is memset and set once, but never checked or used for anything.

It's only used in SpeakNextClause() and related to SelectPhonemeTable().

Another solution void be to keep SelectPhonemeTable() as a void function
and do a current_phoneme_table = voice->phoneme_tab_ix after the call to
SelectPhonemeTable().

global

    Wavegen2() only had one if-statement and then called
    Wavegen(). Wavegen() was never called directly.

Unsigned char *out_start is a global only used in speech.c.

SetSynth_Klatt() was using wvoice in two ways:
1) global wvoice was used directly
2) global wvoice was also passed in Wavegen_Klatt() as an argument "v"

Now the code only uses the function argument wvoice

Contributes to #68.

Contributes to #68.

Wavegen_Klatt2() only had one if-statement and then called
Wavegen_Klatt(). Wavegen_Klatt() was never called directly.

According to [1] and [2] , the pronounciation of the pinyin character
h is not accurately [x].
According to [2], [χ] is instead a possible pronounication in young
users.

As a young native speaker of Mandarin, I feel the sound of [χ] in espeak
more satisfied than [x] here.

[1] https://en.wikipedia.org/wiki/Standard_Chinese_phonology
[2] https://phesoca.com/aws/268/

Signed-off-by: Icenowy Zheng <[email protected]>

This item will lead `冉冉` to be pronounced 3 times rather than 2, and
it's not necessary because this character has no overloaded
pronounciation.

Signed-off-by: Icenowy Zheng <[email protected]>

Allows setting the path for ronn and kramdown

Switch to Portuguese /&~/ for <v>.
Use aveolar /d/ instead of dental /d[/.

The base phonemes file defines alveolar /d/ with the definition for a
dental /d[/ ????