From be07f5e6077e5219b9901ad200b5b7145096a385 Mon Sep 17 00:00:00 2001
From: whp98 <32877126+whp98@users.noreply.github.com>
Date: Mon, 5 Aug 2024 11:08:02 +0800
Subject: [PATCH] update

---
 ...63\350\257\206\345\210\253fast-whisper.md" | 30 +++++++++++++++++++
 ...06\345\214\226\345\244\204\347\220\206.md" | 11 +++++++
 2 files changed, 41 insertions(+)
 create mode 100644 "docs/python\347\233\270\345\205\263/python\350\257\255\351\237\263\350\257\206\345\210\253fast-whisper.md"
 create mode 100644 "docs/\345\205\264\350\266\243/ffmpeg\345\243\260\351\237\263\346\240\207\345\207\206\345\214\226\345\244\204\347\220\206.md"

diff --git "a/docs/python\347\233\270\345\205\263/python\350\257\255\351\237\263\350\257\206\345\210\253fast-whisper.md" "b/docs/python\347\233\270\345\205\263/python\350\257\255\351\237\263\350\257\206\345\210\253fast-whisper.md"
new file mode 100644
index 000000000..b7afea1d6
--- /dev/null
+++ "b/docs/python\347\233\270\345\205\263/python\350\257\255\351\237\263\350\257\206\345\210\253fast-whisper.md"
@@ -0,0 +1,30 @@
+# python语音识别fast-whisper
+
+
+## 语音识别模型下载
+
+```shell
+git clone https://huggingface.co/Systran/faster-whisper-large-v3
+```
+
+
+## 相关使用代码如下
+
+
+```python
+from faster_whisper import WhisperModel
+
+def wisper_generate(audio_path):
+    path = "./faster-whisper-medium"
+    model = WhisperModel(model_size_or_path=path, device="auto", compute_type="int8",
+                         cpu_threads=12)
+    segments, info = model.transcribe(audio=audio_path,
+                                   vad_filter=True,
+                                   vad_parameters=dict(min_silence_duration_ms=1000))
+    print("Detected language '%s' with probability %f" % (info.language, info.language_probability))
+    return segments,info
+if __name__ == "__main__":
+    wisper_generate('./test-cn.mp3')
+    wisper_generate('./test-en.mp3')
+    wisper_generate('./test-ja.wav')
+```
\ No newline at end of file
diff --git "a/docs/\345\205\264\350\266\243/ffmpeg\345\243\260\351\237\263\346\240\207\345\207\206\345\214\226\345\244\204\347\220\206.md" "b/docs/\345\205\264\350\266\243/ffmpeg\345\243\260\351\237\263\346\240\207\345\207\206\345\214\226\345\244\204\347\220\206.md"
new file mode 100644
index 000000000..012d85b17
--- /dev/null
+++ "b/docs/\345\205\264\350\266\243/ffmpeg\345\243\260\351\237\263\346\240\207\345\207\206\345\214\226\345\244\204\347\220\206.md"
@@ -0,0 +1,11 @@
+# ffmpeg-normalize 声音标准化处理
+
+
+## 安装ffmpeg-normalize
+```shell
+pip3 install ffmpeg-normalize
+```
+## 使用ffmpeg-normalize标准化声音到-14LUFS
+```shell
+ffmpeg-normalize ./output_temp.mp3 -o ./output.mp3 -ar 44100 --target -14 --loudness-range 50 --dual-mono -c:a libmp3lame -b:a 192k -pr --dynamic -f
+```
\ No newline at end of file