Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion Jenkinsfile
Original file line number Diff line number Diff line change
Expand Up @@ -28,7 +28,7 @@ pipeline {
MR_TN_CACHE='/home/jenkins/TestData/text_norm/ci/grammars/03-12-24-1'
JA_TN_CACHE='/home/jenkins/TestData/text_norm/ci/grammars/10-17-24-1'
KO_TN_CACHE='/home/jenkins/TestData/text_norm/ci/grammars/04-23-26-0'
HI_TN_CACHE='/home/jenkins/TestData/text_norm/ci/grammars/05-13-26-0'
HI_TN_CACHE='/home/jenkins/TestData/text_norm/ci/grammars/05-18-26-0'
DEFAULT_TN_CACHE='/home/jenkins/TestData/text_norm/ci/grammars/06-08-23-0'
}
stages {
Expand Down
Original file line number Diff line number Diff line change
@@ -0,0 +1,9 @@
₹ रुपया
£ पाउंड
₩ वॉन
$ डॉलर
₺ लीरा
৳ टका
¥ येन
₦ नाइरा
€ यूरो
Original file line number Diff line number Diff line change
@@ -1,9 +1,10 @@
रुपए पैसे
रुपया पैसे
पाउंड पेंस
वॉन जिओन
डॉलर सेंट
लीरा कुरस
टका पैसे
येन सेन
नाइरा कोबो
यूरो सेंट
यूरो सेंट
122 changes: 105 additions & 17 deletions nemo_text_processing/text_normalization/hi/taggers/money.py
Original file line number Diff line number Diff line change
Expand Up @@ -16,18 +16,18 @@
from pynini.lib import pynutil

from nemo_text_processing.text_normalization.hi.graph_utils import GraphFst, insert_space
from nemo_text_processing.text_normalization.hi.utils import get_abs_path
from nemo_text_processing.text_normalization.hi.utils import get_abs_path, load_labels

currency_graph = pynini.string_file(get_abs_path("data/money/currency.tsv"))
currency_singular_graph = pynini.string_file(get_abs_path("data/money/currency_singular.tsv"))


class MoneyFst(GraphFst):
"""
Finite state transducer for classifying money, suppletive aware, e.g.
₹५० -> money { money { currency_maj: "रुपए" integer_part: "पचास" }
₹५०.५० -> money { currency_maj: "रुपए" integer_part: "पचास" fractional_part: "पचास" currency_min: "centiles" }
₹०.५० -> money { currency_maj: "रुपए" integer_part: "शून्य" fractional_part: "पचास" currency_min: "centiles" }
Note that the 'centiles' string is a placeholder to handle by the verbalizer by applying the corresponding minor currency denomination
₹५० -> money { currency_maj: "रुपए" integer_part: "पचास" }
₹५०.५० -> money { currency_maj: "रुपए" integer_part: "पचास" fractional_part: "पचास" currency_min: "पैसे" }
₹०.५० -> { money { currency_maj: "रुपए" integer_part: "शून्य" fractional_part: "पचास" currency_min: "पैसे" }

Args:
cardinal: CardinalFst
Expand All @@ -41,30 +41,118 @@ def __init__(self, cardinal: GraphFst):

cardinal_graph = cardinal.final_graph

_en_to_hi_digit = pynini.string_file(get_abs_path("data/ordinal/en_to_hi_digit.tsv"))
_deva_to_ascii = pynini.invert(_en_to_hi_digit)
deva_to_ascii = pynini.closure(_deva_to_ascii | pynini.union(*"0123456789"), 1)

_ascii_digit = pynini.union(*"0123456789")
_ascii_nonzero = pynini.union(*"123456789")
_deva_nonzero = pynini.union(*"१२३४५६७८९")
_any_digit = _ascii_digit | pynini.union(*"०१२३४५६७८९")
_any_nonzero = _ascii_nonzero | _deva_nonzero

optional_graph_negative = pynini.closure(
pynutil.insert("negative: ") + pynini.cross("-", "\"true\"") + insert_space,
pynutil.insert("negative: ") + pynini.cross("-", '"true"') + insert_space,
0,
1,
)

currency_major = pynutil.insert('currency_maj: "') + currency_graph + pynutil.insert('"')
currency_major_singular = pynutil.insert('currency_maj: "') + currency_singular_graph + pynutil.insert('"')

one = pynini.union("1", "१")
integer_one = pynutil.insert('integer_part: "') + (one @ cardinal_graph) + pynutil.insert('"')
integer = pynutil.insert('integer_part: "') + cardinal_graph + pynutil.insert('"')
fraction = pynutil.insert('fractional_part: "') + cardinal_graph + pynutil.insert('"')
currency_minor = pynutil.insert('currency_min: "') + pynutil.insert("centiles") + pynutil.insert('"')

graph_major_only = optional_graph_negative + currency_major + insert_space + integer
graph_major_and_minor = (
strip_trailing_zeros = pynini.closure(_ascii_digit) + _ascii_nonzero + pynini.closure(pynutil.delete("0"))
canonicalise = (
(pynutil.delete("0") + _ascii_nonzero)
| (_ascii_nonzero + pynutil.insert("0"))
| (_ascii_nonzero + _ascii_digit)
)
two_digits_fractional_part = deva_to_ascii @ strip_trailing_zeros @ canonicalise

fraction = (
pynutil.insert('fractional_part: "') + (two_digits_fractional_part @ cardinal_graph) + pynutil.insert('"')
)

optional_delete_fractional_zeros = pynini.closure(
pynutil.delete(".") + pynini.closure(pynutil.delete("0") | pynutil.delete("०"), 1),
0,
1,
)

has_3plus_sig_digits = _any_digit + _any_digit + _any_nonzero + pynini.closure(_any_digit)
single_digit = _any_digit @ cardinal.single_digits_graph
decimal_digits = (
pynutil.insert('fractional_part: "')
+ single_digit
+ pynini.closure(insert_space + single_digit)
+ pynutil.insert('"')
)
guarded_decimal_digits = has_3plus_sig_digits @ decimal_digits

graph_decimal_path = (
optional_graph_negative
+ currency_major
+ insert_space
+ integer
+ pynutil.insert('integer_part: "')
+ cardinal_graph
+ pynutil.insert('"')
+ pynini.cross(".", " ")
+ fraction
+ guarded_decimal_digits
).optimize()

graph_major_only_singular = (
optional_graph_negative
+ currency_major_singular
+ insert_space
+ currency_minor
)
+ integer_one
+ optional_delete_fractional_zeros
).optimize()

graph_major_only = (
optional_graph_negative + currency_major + insert_space + integer + optional_delete_fractional_zeros
).optimize()

maj_labels = load_labels(get_abs_path("data/money/currency.tsv"))
maj_singular_labels = load_labels(get_abs_path("data/money/currency_singular.tsv"))
maj_to_min = dict(load_labels(get_abs_path("data/money/major_minor_currencies.tsv")))

def _build_major_and_minor(sym_maj_labels, int_graph):
result = None
for sym, maj in sym_maj_labels:
min_name = maj_to_min.get(maj)
if not min_name:
continue

graph_currencies = graph_major_only | graph_major_and_minor
curr_maj = pynutil.insert('currency_maj: "') + pynini.cross(sym, maj) + pynutil.insert('"')
curr_min = pynutil.insert('currency_min: "') + pynutil.insert(min_name) + pynutil.insert('"')

g = (
optional_graph_negative
+ curr_maj
+ insert_space
+ int_graph
+ pynini.cross(".", " ")
+ fraction
+ insert_space
+ curr_min
).optimize()

result = g if result is None else pynini.union(result, g).optimize()

return result

graph_major_and_minor = _build_major_and_minor(maj_labels, integer)
graph_major_and_minor_singular = _build_major_and_minor(maj_singular_labels, integer_one)

graph_currencies = (
pynutil.add_weight(graph_major_only_singular | graph_major_and_minor_singular, -0.001)
| pynutil.add_weight(graph_decimal_path, -0.0005)
| graph_major_only
| graph_major_and_minor
)

graph = graph_currencies.optimize()
final_graph = self.add_tokens(graph)
self.fst = final_graph
self.fst = self.add_tokens(graph)
98 changes: 50 additions & 48 deletions nemo_text_processing/text_normalization/hi/verbalizers/money.py
Original file line number Diff line number Diff line change
Expand Up @@ -15,26 +15,16 @@
import pynini
from pynini.lib import pynutil

major_minor_currencies = {
"रुपए": "पैसे",
"पाउंड": "पेंस",
"वॉन": "जिओन",
"डॉलर": "सेंट",
"लीरा": "कुरस",
"टका": "पैसे",
"येन": "सेन",
"नाइरा": "कोबो",
"यूरो": "सेंट",
}
from nemo_text_processing.text_normalization.hi.graph_utils import NEMO_NOT_QUOTE, NEMO_SPACE, GraphFst
from nemo_text_processing.text_normalization.hi.utils import get_abs_path, load_labels


class MoneyFst(GraphFst):
"""
Finite state transducer for verbalizing money, e.g.
money { integer_part: "बारह" currency_maj: "रुपए" } -> बारह रुपए
money { integer_part: "बारह" currency_maj: "रुपए" fractional_part: "पचास" currency_min: "centiles" } -> बारह रुपए पचास पैसे
money { currency_maj: "रुपए" integer_part: "शून्य" fractional_part: "पचास" currency_min: "centiles" } -> पचास पैसे
money { currency_maj: "रुपए" integer_part: "बारह" } } -> बारह रुपए
money { currency_maj: "रुपए" integer_part: "बारह" fractional_part: "पचास" currency_min: "पैसे" } -> बारह रुपए पचास पैसे
money { currency_maj: "रुपए" integer_part: "शून्य" fractional_part: "पचास" currency_min: "पैसे" } -> पचास पैसे

Args:
cardinal: CardinalFst
Expand All @@ -46,55 +36,67 @@ class MoneyFst(GraphFst):
def __init__(self):
super().__init__(name="money", kind="verbalize")

currency_major = pynutil.delete('currency_maj: "') + pynini.closure(NEMO_NOT_QUOTE, 1) + pynutil.delete('"')
sp = pynini.accep(NEMO_SPACE)

currency_major = pynutil.delete('currency_maj: "') + pynini.closure(NEMO_NOT_QUOTE, 1) + pynutil.delete('"')
integer_part = pynutil.delete('integer_part: "') + pynini.closure(NEMO_NOT_QUOTE, 1) + pynutil.delete('"')

fractional_part = (
pynutil.delete('fractional_part: "') + pynini.closure(NEMO_NOT_QUOTE, 1) + pynutil.delete('"')
)

# Handles major denominations only
graph_major_only = integer_part + pynini.accep(NEMO_SPACE) + currency_major
currency_minor = pynutil.delete('currency_min: "') + pynini.closure(NEMO_NOT_QUOTE, 1) + pynutil.delete('"')

graph_major_only = integer_part + sp + currency_major

all_major_names = [maj for maj, _ in load_labels(get_abs_path("data/money/major_minor_currencies.tsv"))]

# Handles both major and minor denominations
major_minor_graphs = []
minor_only_graphs = []

for major in all_major_names:
graph_major_slot = pynutil.delete('currency_maj: "') + pynutil.delete(major) + pynutil.delete('"')

major_minor_graphs.append(
graph_major_slot
+ sp
+ integer_part
+ pynutil.insert(NEMO_SPACE)
+ pynutil.insert(major)
+ sp
+ fractional_part
+ sp
+ currency_minor
)

# Handles minor denominations only
minor_graphs = []

# Logic for handling minor denominations
for major, minor in major_minor_currencies.items():
graph_major = pynutil.delete('currency_maj: "') + pynini.accep(major) + pynutil.delete('"')
graph_minor = pynutil.delete('currency_min: "') + pynini.cross("centiles", minor) + pynutil.delete('"')
graph_major_minor_partial = (
integer_part
+ pynini.accep(NEMO_SPACE)
+ graph_major
+ pynini.accep(NEMO_SPACE)
minor_only_graphs.append(
graph_major_slot
+ sp
+ pynutil.delete('integer_part: "शून्य"')
+ sp
+ fractional_part
+ pynini.accep(NEMO_SPACE)
+ graph_minor
+ sp
+ currency_minor
)
major_minor_graphs.append(graph_major_minor_partial)

graph_minor_partial = (
pynutil.delete('integer_part: "शून्य"')
+ pynutil.delete(NEMO_SPACE)
+ pynutil.delete('currency_maj: "')
graph_major_minor = pynini.union(*major_minor_graphs)
graph_minor_only = pynini.union(*minor_only_graphs)

decimal_graphs = []
for major in all_major_names:
decimal_graphs.append(
pynutil.delete('currency_maj: "')
+ pynutil.delete(major)
+ pynutil.delete('"')
+ pynutil.delete(NEMO_SPACE)
+ sp
+ integer_part
+ sp
+ pynutil.insert(" दशमलव ")
+ fractional_part
+ pynini.accep(NEMO_SPACE)
+ graph_minor
+ pynutil.insert(NEMO_SPACE)
+ pynutil.insert(major)
)
minor_graphs.append(graph_minor_partial)

graph_major_minor = pynini.union(*major_minor_graphs)
graph_minor_only = pynini.union(*minor_graphs)
graph_decimal_money = pynini.union(*decimal_graphs)

graph = graph_major_only | graph_major_minor | pynutil.add_weight(graph_minor_only, -0.1)
graph = graph_major_only | graph_major_minor | pynutil.add_weight(graph_minor_only, -0.1) | graph_decimal_money

delete_tokens = self.delete_tokens(graph)
self.fst = delete_tokens.optimize()
self.fst = self.delete_tokens(graph).optimize()
Original file line number Diff line number Diff line change
Expand Up @@ -116,3 +116,11 @@ $९.९९~नौ डॉलर निन्यानबे सेंट
₦१०.२७~दस नाइरा सत्ताईस कोबो
€200.90~दो सौ यूरो नब्बे सेंट
€१२३४.७५~एक हज़ार दो सौ चौंतीस यूरो पचहत्तर सेंट
$1.12~एक डॉलर बारह सेंट
$1.123~एक दशमलव एक दो तीन डॉलर
$1.1234~एक दशमलव एक दो तीन चार डॉलर
₹2.2000~दो रुपए बीस पैसे
$1.2000~एक डॉलर बीस सेंट
₹1.500~एक रुपया पचास पैसे
₹5.00~पाँच रुपए
₹१~एक रुपया