41 lines
1.3 KiB
Python
41 lines
1.3 KiB
Python
import pynini
|
|
from fun_text_processing.text_normalization.en.graph_utils import (
|
|
DAMO_NOT_QUOTE,
|
|
GraphFst,
|
|
delete_preserve_order,
|
|
)
|
|
from pynini.lib import pynutil
|
|
|
|
|
|
class TelephoneFst(GraphFst):
|
|
"""
|
|
Finite state transducer for verbalizing telephone, e.g.
|
|
telephone { country_code: "plus neun und vierzig" number_part: "null eins eins eins null null null" }
|
|
-> "plus neun und vierzig null eins eins eins null null null"
|
|
|
|
Args:
|
|
deterministic: if True will provide a single transduction option,
|
|
for False multiple transduction are generated (used for audio-based normalization)
|
|
"""
|
|
|
|
def __init__(self, deterministic: bool = True):
|
|
super().__init__(name="telephone", kind="verbalize", deterministic=deterministic)
|
|
|
|
country_code = (
|
|
pynutil.delete('country_code: "')
|
|
+ pynini.closure(DAMO_NOT_QUOTE, 1)
|
|
+ pynutil.delete('"')
|
|
)
|
|
|
|
number_part = (
|
|
pynutil.delete('number_part: "')
|
|
+ pynini.closure(DAMO_NOT_QUOTE, 1)
|
|
+ pynutil.delete('"')
|
|
)
|
|
|
|
self.graph = country_code + pynini.accep(" ") + number_part
|
|
|
|
graph = self.graph + delete_preserve_order
|
|
delete_tokens = self.delete_tokens(graph)
|
|
self.fst = delete_tokens.optimize()
|