From 9f6445d39b14fa17f2c32a383c1054a8e073a9c9 Mon Sep 17 00:00:00 2001
From: 游雁 <zhifu.gzf@alibaba-inc.com>
Date: 星期五, 07 四月 2023 14:11:55 +0800
Subject: [PATCH] onnx
---
funasr/export/models/vad_realtime_transformer.py | 37 +++++++++++++++++++++++--------------
1 files changed, 23 insertions(+), 14 deletions(-)
diff --git a/funasr/export/models/vad_realtime_transformer.py b/funasr/export/models/vad_realtime_transformer.py
index a3d4864..24a8e72 100644
--- a/funasr/export/models/vad_realtime_transformer.py
+++ b/funasr/export/models/vad_realtime_transformer.py
@@ -1,14 +1,9 @@
-from typing import Any
-from typing import List
from typing import Tuple
import torch
import torch.nn as nn
-from funasr.modules.embedding import SinusoidalPositionEncoder
-from funasr.punctuation.sanm_encoder import SANMVadEncoder as Encoder
-from funasr.punctuation.abs_model import AbsPunctuation
-from funasr.punctuation.sanm_encoder import SANMVadEncoder
+from funasr.models.encoder.sanm_encoder import SANMVadEncoder
from funasr.export.models.encoder.sanm_encoder import SANMVadEncoder as SANMVadEncoder_export
class VadRealtimeTransformer(nn.Module):
@@ -57,17 +52,27 @@
def with_vad(self):
return True
- def get_dummy_inputs(self):
- length = 120
- text_indexes = torch.randint(0, self.embed.num_embeddings, (1, length))
+ # def get_dummy_inputs(self):
+ # length = 120
+ # text_indexes = torch.randint(0, self.embed.num_embeddings, (1, length))
+ # text_lengths = torch.tensor([length], dtype=torch.int32)
+ # vad_mask = torch.ones(length, length, dtype=torch.float32)[None, None, :, :]
+ # sub_masks = torch.ones(length, length, dtype=torch.float32)
+ # sub_masks = torch.tril(sub_masks).type(torch.float32)
+ # return (text_indexes, text_lengths, vad_mask, sub_masks[None, None, :, :])
+
+ def get_dummy_inputs(self, txt_dir=None):
+ from funasr.modules.mask import vad_mask
+ length = 10
+ text_indexes = torch.tensor([[266757, 266757, 266757, 266757, 266757, 266757, 266757, 266757, 266757, 266757]], dtype=torch.int32)
text_lengths = torch.tensor([length], dtype=torch.int32)
- vad_mask = torch.ones(length, length, dtype=torch.float32)[None, None, :, :]
+ vad_masks = vad_mask(10, 2, dtype=torch.float32)[None, None, :, :]
sub_masks = torch.ones(length, length, dtype=torch.float32)
- sub_masks = torch.tril(sub_masks)
- return (text_indexes, text_lengths, vad_mask, sub_masks)
+ sub_masks = torch.tril(sub_masks).type(torch.float32)
+ return (text_indexes, text_lengths, vad_masks, sub_masks[None, None, :, :])
def get_input_names(self):
- return ['input', 'text_lengths', 'vad_mask']
+ return ['input', 'text_lengths', 'vad_masks', 'sub_masks']
def get_output_names(self):
return ['logits']
@@ -77,7 +82,11 @@
'input': {
1: 'feats_length'
},
- 'vad_mask': {
+ 'vad_masks': {
+ 2: 'feats_length1',
+ 3: 'feats_length2'
+ },
+ 'sub_masks': {
2: 'feats_length1',
3: 'feats_length2'
},
--
Gitblit v1.9.1