From 1596f6f414f6f41da66506debb1dff19fffeb3ec Mon Sep 17 00:00:00 2001
From: 游雁 <zhifu.gzf@alibaba-inc.com>
Date: 星期一, 24 六月 2024 11:55:17 +0800
Subject: [PATCH] fixbug hotwords

---
 examples/industrial_data_pretraining/fsmn_vad_streaming/demo.py |   42 +++++++++++++++++++++++++++---------------
 1 files changed, 27 insertions(+), 15 deletions(-)

diff --git a/examples/industrial_data_pretraining/fsmn_vad_streaming/demo.py b/examples/industrial_data_pretraining/fsmn_vad_streaming/demo.py
index 6761a80..21ce0cb 100644
--- a/examples/industrial_data_pretraining/fsmn_vad_streaming/demo.py
+++ b/examples/industrial_data_pretraining/fsmn_vad_streaming/demo.py
@@ -4,16 +4,16 @@
 #  MIT License  (https://opensource.org/licenses/MIT)
 
 from funasr import AutoModel
+
 wav_file = "https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ASR/test_audio/vad_example.wav"
 
-chunk_size = 60000 # ms
-model = AutoModel(model="/Users/zhifu/Downloads/modelscope_models/speech_fsmn_vad_zh-cn-16k-common-streaming", model_revision="v2.0.0")
+model = AutoModel(model="iic/speech_fsmn_vad_zh-cn-16k-common-pytorch")
 
-res = model(input=wav_file,
-            chunk_size=chunk_size,
-            )
+res = model.generate(input=wav_file)
 print(res)
 
+# [[beg1, end1], [beg2, end2], .., [begN, endN]]
+# beg/end: ms
 
 
 import soundfile
@@ -22,16 +22,28 @@
 wav_file = os.path.join(model.model_path, "example/vad_example.wav")
 speech, sample_rate = soundfile.read(wav_file)
 
-chunk_stride = int(chunk_size * 16000 / 1000)
+chunk_size = 200  # ms
+chunk_stride = int(chunk_size * sample_rate / 1000)
 
 cache = {}
 
-for i in range(int(len((speech)-1)/chunk_stride+1)):
-    speech_chunk = speech[i*chunk_stride:(i+1)*chunk_stride]
-    is_final = i == int(len((speech)-1)/chunk_stride+1) - 1
-    res = model(input=speech_chunk,
-                cache=cache,
-                is_final=is_final,
-                chunk_size=chunk_size,
-                )
-    print(res)
+total_chunk_num = int(len((speech) - 1) / chunk_stride + 1)
+for i in range(total_chunk_num):
+    speech_chunk = speech[i * chunk_stride : (i + 1) * chunk_stride]
+    is_final = i == total_chunk_num - 1
+    res = model.generate(
+        input=speech_chunk,
+        cache=cache,
+        is_final=is_final,
+        chunk_size=chunk_size,
+        disable_pbar=True,
+    )
+    # print(res)
+    if len(res[0]["value"]):
+        print(res)
+
+
+# 1. [[beg1, end1], [beg2, end2], .., [begN, endN]]; [[beg, end]]; [[beg1, end1], [beg2, end2]]
+# 2. [[beg, -1]]
+# 3. [[-1, end]]
+# beg/end: ms

--
Gitblit v1.9.1