From 9c0735b7dfb1eaa1d8f186fe99f7d29e04a1178b Mon Sep 17 00:00:00 2001
From: 游雁 <zhifu.gzf@alibaba-inc.com>
Date: 星期五, 22 三月 2024 19:09:49 +0800
Subject: [PATCH] update
---
examples/README_zh.md | 131 ++++++++++++++++++++++++++++++++++++++++++-
1 files changed, 128 insertions(+), 3 deletions(-)
diff --git a/examples/README_zh.md b/examples/README_zh.md
index 0e3846d..59d1303 100644
--- a/examples/README_zh.md
+++ b/examples/README_zh.md
@@ -16,17 +16,18 @@
print(res)
```
-### 璇︾粏鐢ㄦ硶浠嬬粛
+### 鎺ュ彛璇存槑
#### AutoModel 瀹氫箟
```python
-model = AutoModel(model=[str], device=[str], ncpu=[int], output_dir=[str], batch_size=[int], **kwargs)
+model = AutoModel(model=[str], device=[str], ncpu=[int], output_dir=[str], batch_size=[int], hub=[str], **kwargs)
```
- `model`(str): [妯″瀷浠撳簱](https://github.com/alibaba-damo-academy/FunASR/tree/main/model_zoo) 涓殑妯″瀷鍚嶇О锛屾垨鏈湴纾佺洏涓殑妯″瀷璺緞
- `device`(str): `cuda:0`锛堥粯璁pu0锛夛紝浣跨敤 GPU 杩涜鎺ㄧ悊锛屾寚瀹氥�傚鏋滀负`cpu`锛屽垯浣跨敤 CPU 杩涜鎺ㄧ悊
- `ncpu`(int): `4` 锛堥粯璁わ級锛岃缃敤浜� CPU 鍐呴儴鎿嶄綔骞惰鎬х殑绾跨▼鏁�
- `output_dir`(str): `None` 锛堥粯璁わ級锛屽鏋滆缃紝杈撳嚭缁撴灉鐨勮緭鍑鸿矾寰�
-- `batch_size`(int): `1` 锛堥粯璁わ級锛岃В鐮佹椂鐨勬壒澶勭悊澶у皬
+- `batch_size`(int): `1` 锛堥粯璁わ級锛岃В鐮佹椂鐨勬壒澶勭悊锛屾牱鏈釜鏁�
+- `hub`(str)锛歚ms`锛堥粯璁わ級锛屼粠modelscope涓嬭浇妯″瀷銆傚鏋滀负`hf`锛屼粠huggingface涓嬭浇妯″瀷銆�
- `**kwargs`(dict): 鎵�鏈夊湪`config.yaml`涓弬鏁帮紝鍧囧彲浠ョ洿鎺ュ湪姝ゅ鎸囧畾锛屼緥濡傦紝vad妯″瀷涓渶澶у垏鍓查暱搴� `max_single_segment_time=6000` 锛堟绉掞級銆�
#### AutoModel 鎺ㄧ悊
@@ -57,6 +58,130 @@
- `type`(str)锛歚onnx`(榛樿)锛屽鍑簅nnx鏍煎紡銆俙torch`瀵煎嚭libtorch鏍煎紡銆�
- `quantize`(bool)锛歚False`锛堥粯璁わ級锛屾槸鍚﹀仛閲忓寲銆�
+### 鏇村鐢ㄦ硶浠嬬粛
+
+
+#### 闈炲疄鏃惰闊宠瘑鍒�
+```python
+from funasr import AutoModel
+# paraformer-zh is a multi-functional asr model
+# use vad, punc, spk or not as you need
+model = AutoModel(model="paraformer-zh",
+ vad_model="fsmn-vad",
+ vad_kwargs={"max_single_segment_time": 60000},
+ punc_model="ct-punc",
+ # spk_model="cam++"
+ )
+wav_file = f"{model.model_path}/example/asr_example.wav"
+res = model.generate(input=wav_file, batch_size_s=300, batch_size_threshold_s=60, hotword='榄旀惌')
+print(res)
+```
+娉ㄦ剰锛�
+- 閫氬父妯″瀷杈撳叆闄愬埗鏃堕暱30s浠ヤ笅锛岀粍鍚坄vad_model`鍚庯紝鏀寔浠绘剰鏃堕暱闊抽杈撳叆锛屼笉灞�闄愪簬paraformer妯″瀷锛屾墍鏈夐煶棰戣緭鍏ユā鍨嬪潎鍙互銆�
+- `model`鐩稿叧鐨勫弬鏁板彲浠ョ洿鎺ュ湪`AutoModel`瀹氫箟涓洿鎺ユ寚瀹氾紱涓巂vad_model`鐩稿叧鍙傛暟鍙互閫氳繃`vad_kwargs`鏉ユ寚瀹氾紝绫诲瀷涓篸ict锛涚被浼肩殑鏈塦punc_kwargs`锛宍spk_kwargs`锛�
+- `max_single_segment_time`: 琛ㄧず`vad_model`鏈�澶у垏鍓查煶棰戞椂闀�, 鍗曚綅鏄绉抦s.
+- `batch_size_s` 琛ㄧず閲囩敤鍔ㄦ�乥atch锛宐atch涓�婚煶棰戞椂闀匡紝鍗曚綅涓虹s銆�
+- `batch_size_threshold_s`: 琛ㄧず`vad_model`鍒囧壊鍚庨煶棰戠墖娈垫椂闀胯秴杩� `batch_size_threshold_s`闃堝�兼椂锛屽皢batch_size鏁拌缃负1, 鍗曚綅涓虹s.
+
+寤鸿锛氬綋鎮ㄨ緭鍏ヤ负闀块煶棰戯紝閬囧埌OOM闂鏃讹紝鍥犱负鏄惧瓨鍗犵敤涓庨煶棰戞椂闀垮憟骞虫柟鍏崇郴澧炲姞锛屽垎涓�3绉嶆儏鍐碉細
+- a)鎺ㄧ悊璧峰闃舵锛屾樉瀛樹富瑕佸彇鍐充簬`batch_size_s`锛岄�傚綋鍑忓皬璇ュ�硷紝鍙互鍑忓皯鏄惧瓨鍗犵敤锛�
+- b)鎺ㄧ悊涓棿闃舵锛岄亣鍒癡AD鍒囧壊鐨勯暱闊抽鐗囨锛屾�籺oken鏁板皬浜巂batch_size_s`锛屼粛鐒跺嚭鐜癘OM锛屽彲浠ラ�傚綋鍑忓皬`batch_size_threshold_s`锛岃秴杩囬槇鍊硷紝寮哄埗batch涓�1;
+- c)鎺ㄧ悊蹇粨鏉熼樁娈碉紝閬囧埌VAD鍒囧壊鐨勯暱闊抽鐗囨锛屾�籺oken鏁板皬浜巂batch_size_s`锛屼笖瓒呰繃闃堝�糮batch_size_threshold_s`锛屽己鍒禸atch涓�1锛屼粛鐒跺嚭鐜癘OM锛屽彲浠ラ�傚綋鍑忓皬`max_single_segment_time`锛屼娇寰梀AD鍒囧壊闊抽鏃堕暱鍙樼煭銆�
+
+#### 瀹炴椂璇煶璇嗗埆
+
+```python
+from funasr import AutoModel
+
+chunk_size = [0, 10, 5] #[0, 10, 5] 600ms, [0, 8, 4] 480ms
+encoder_chunk_look_back = 4 #number of chunks to lookback for encoder self-attention
+decoder_chunk_look_back = 1 #number of encoder chunks to lookback for decoder cross-attention
+
+model = AutoModel(model="paraformer-zh-streaming")
+
+import soundfile
+import os
+
+wav_file = os.path.join(model.model_path, "example/asr_example.wav")
+speech, sample_rate = soundfile.read(wav_file)
+chunk_stride = chunk_size[1] * 960 # 600ms
+
+cache = {}
+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, encoder_chunk_look_back=encoder_chunk_look_back, decoder_chunk_look_back=decoder_chunk_look_back)
+ print(res)
+```
+
+娉細`chunk_size`涓烘祦寮忓欢鏃堕厤缃紝`[0,10,5]`琛ㄧず涓婂睆瀹炴椂鍑哄瓧绮掑害涓篳10*60=600ms`锛屾湭鏉ヤ俊鎭负`5*60=300ms`銆傛瘡娆℃帹鐞嗚緭鍏ヤ负`600ms`锛堥噰鏍风偣鏁颁负`16000*0.6=960`锛夛紝杈撳嚭涓哄搴旀枃瀛楋紝鏈�鍚庝竴涓闊崇墖娈佃緭鍏ラ渶瑕佽缃甡is_final=True`鏉ュ己鍒惰緭鍑烘渶鍚庝竴涓瓧銆�
+
+#### 璇煶绔偣妫�娴嬶紙闈炲疄鏃讹級
+```python
+from funasr import AutoModel
+
+model = AutoModel(model="fsmn-vad")
+
+wav_file = f"{model.model_path}/example/asr_example.wav"
+res = model.generate(input=wav_file)
+print(res)
+```
+娉細VAD妯″瀷杈撳嚭鏍煎紡涓猴細`[[beg1, end1], [beg2, end2], .., [begN, endN]]`锛屽叾涓璥begN/endN`琛ㄧず绗琡N`涓湁鏁堥煶棰戠墖娈电殑璧峰鐐�/缁撴潫鐐癸紝
+鍗曚綅涓烘绉掋��
+
+#### 璇煶绔偣妫�娴嬶紙瀹炴椂锛�
+```python
+from funasr import AutoModel
+
+chunk_size = 200 # ms
+model = AutoModel(model="fsmn-vad")
+
+import soundfile
+
+wav_file = f"{model.model_path}/example/vad_example.wav"
+speech, sample_rate = soundfile.read(wav_file)
+chunk_stride = int(chunk_size * sample_rate / 1000)
+
+cache = {}
+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)
+ if len(res[0]["value"]):
+ print(res)
+```
+娉細娴佸紡VAD妯″瀷杈撳嚭鏍煎紡涓�4绉嶆儏鍐碉細
+- `[[beg1, end1], [beg2, end2], .., [begN, endN]]`锛氬悓涓婄绾縑AD杈撳嚭缁撴灉銆�
+- `[[beg, -1]]`锛氳〃绀哄彧妫�娴嬪埌璧峰鐐广��
+- `[[-1, end]]`锛氳〃绀哄彧妫�娴嬪埌缁撴潫鐐广��
+- `[]`锛氳〃绀烘棦娌℃湁妫�娴嬪埌璧峰鐐癸紝涔熸病鏈夋娴嬪埌缁撴潫鐐�
+杈撳嚭缁撴灉鍗曚綅涓烘绉掞紝浠庤捣濮嬬偣寮�濮嬬殑缁濆鏃堕棿銆�
+
+#### 鏍囩偣鎭㈠
+```python
+from funasr import AutoModel
+
+model = AutoModel(model="ct-punc")
+
+res = model.generate(input="閭ｄ粖澶╃殑浼氬氨鍒拌繖閲屽惂 happy new year 鏄庡勾瑙�")
+print(res)
+```
+
+#### 鏃堕棿鎴抽娴�
+```python
+from funasr import AutoModel
+
+model = AutoModel(model="fa-zh")
+
+wav_file = f"{model.model_path}/example/asr_example.wav"
+text_file = f"{model.model_path}/example/text.txt"
+res = model.generate(input=(wav_file, text_file), data_type=("sound", "text"))
+print(res)
+```
+鏇村锛圼绀轰緥](https://github.com/alibaba-damo-academy/FunASR/tree/main/examples/industrial_data_pretraining)锛�
+
## 寰皟
--
Gitblit v1.9.1