SAMPA to mbrola phoneme conversion

[alpoktem]

I need to get the phonetic transcriptions of words that are in my dictionary with the same phoneme set used in the pho files. I am working with the mb-es1 voice.

espeak has an interface to obtain phonetic transcription but it doesn't give me the same transcription as it is seen in the pho file it produces. The phoneme set that the mbrola voice uses seems different and much simpler than the SAMPA transcription that espeak outputs for the voice.

For the word "querer", when I use the -x option of espeak I get the following transcription:
kE**'Er

However, in the pho output I see the transcription as "kerer".

k	102
e	35	 0 97 20 97 40 98 59 98 80 98 100 98
r	65
e	85	 0 104 13 93 27 81 53 90 80 106 100 106
r	65
_	150

I need to get the transcription as mbrola uses. I'd really appreciate if you could help.

[bphd]

Searched for something generating phoneme, got one here, but got the fact that developers of MBROLA would not help about that part at all too. Sad

[valdisvi]

https://github.com/espeak-ng/espeak-ng/blob/master/docs/mbrola.md