Open-source speech recognition toolkit for training, inference, streaming ASR, VAD, punctuation, speaker diarization pipelines, and OpenAI-compatible/MCP serving.
Python asraudiochineseemotion-recognitionfunasrmcp-servermultilingual-asropenai-compatible-apiparaformerpunctuationpytorchreal-time-asrspeaker-diarizationspeech-recognitionspeech-to-textstreaming-asrtranscriptionvllmvoice-activity-detectionwhisper-alternative
import pynini
from fun_text_processing.text_normalization.en.graph_utils import (
DAMO_NOT_QUOTE,
GraphFst,
delete_space,
)
from pynini.lib import pynutil
class CardinalFst(GraphFst):
"""
Finite state transducer for verbalizing cardinal
e.g. cardinal { integer: "23" negative: "-" } -> -23
"""
def __init__(self):
super().__init__(name="cardinal", kind="verbalize")
optional_sign = pynini.closure(
pynutil.delete("negative:")
+ delete_space
+ pynutil.delete('"')
+ DAMO_NOT_QUOTE
+ pynutil.delete('"')
+ delete_space,
0,
1,
)
graph = (
pynutil.delete("integer:")
+ delete_space
+ pynutil.delete('"')
+ pynini.closure(DAMO_NOT_QUOTE, 1)
+ pynutil.delete('"')
)
self.numbers = graph
graph = optional_sign + graph
delete_tokens = self.delete_tokens(graph)
self.fst = delete_tokens.optimize()