podcast-splitter

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Podcast Splitter

播客分割工具

Automatically split audio files into segments based on silence detection. Perfect for dividing podcasts into chapters, creating clips from long recordings, or removing dead air.

基于静音检测自动将音频文件分割为多个片段。非常适合将播客划分为章节、从长录音中创建片段或移除空白静音部分。

Quick Start

快速开始

python

from scripts.podcast_splitter import PodcastSplitter

python

from scripts.podcast_splitter import PodcastSplitter

Auto-split by silence

按静音自动分割

splitter = PodcastSplitter("podcast_episode.mp3") segments = splitter.split_by_silence() splitter.export_segments("./chapters/")

Remove long silences

移除长静音

splitter = PodcastSplitter("raw_recording.mp3") splitter.remove_silence(min_length=2000) # Remove silences > 2 seconds splitter.save("clean_recording.mp3")

undefined

splitter = PodcastSplitter("raw_recording.mp3") splitter.remove_silence(min_length=2000) # 移除时长超过2秒的静音 splitter.save("clean_recording.mp3")

undefined

Features

功能特性

Silence Detection: Configurable threshold and duration
Auto-Split: Divide audio at natural breaks
Silence Removal: Remove or shorten long pauses
Chapter Export: Save individual segments as files
Preview Mode: List detected silences without splitting
Batch Processing: Process multiple files

静音检测：可配置阈值和时长
自动分割：在自然停顿处分割音频
静音移除：移除或缩短长停顿
章节导出：将单个片段保存为文件
预览模式：列出检测到的静音而不进行分割
批量处理：处理多个文件

API Reference

API 参考

Initialization

初始化

python

splitter = PodcastSplitter("audio.mp3")

python

splitter = PodcastSplitter("audio.mp3")

With custom settings

自定义设置

splitter = PodcastSplitter( "audio.mp3", silence_thresh=-40, # dBFS threshold min_silence_len=1000, # Minimum silence length (ms) keep_silence=300 # Silence to keep at segment edges (ms) )

undefined

splitter = PodcastSplitter( "audio.mp3", silence_thresh=-40, # dBFS阈值 min_silence_len=1000, # 最小静音时长（毫秒） keep_silence=300 # 片段边缘保留的静音时长（毫秒） )

undefined

Silence Detection

静音检测

python

undefined

python

undefined

Detect silence regions

检测静音区域

silences = splitter.detect_silence()

Returns: [(start_ms, end_ms), (start_ms, end_ms), ...]

返回：[(开始毫秒, 结束毫秒), (开始毫秒, 结束毫秒), ...]

Print silence summary

打印静音摘要

splitter.print_silence_report()

undefined

splitter.print_silence_report()

undefined

Splitting

分割操作

python

undefined

python

undefined

Split at all detected silences

在所有检测到的静音处分割

segments = splitter.split_by_silence()

Split at silences longer than threshold

在长于指定时长的静音处分割

segments = splitter.split_by_silence(min_silence_len=3000)

Limit number of segments

限制片段数量

segments = splitter.split_by_silence(max_segments=10)

undefined

segments = splitter.split_by_silence(max_segments=10)

undefined

Silence Removal

静音移除

python

undefined

python

undefined

Remove silences longer than threshold

移除长于指定时长的静音

splitter.remove_silence(min_length=2000) # Remove >2s silences

splitter.remove_silence(min_length=2000) # 移除时长>2秒的静音

Shorten silences to max length

将静音缩短至最大时长

splitter.shorten_silence(max_length=500) # Cap at 500ms

splitter.shorten_silence(max_length=500) # 静音时长上限为500毫秒

Remove leading/trailing silence only

仅移除开头/结尾的静音

splitter.strip_silence()

undefined

splitter.strip_silence()

undefined

Export

导出操作

python

undefined

python

undefined

Export all segments

导出所有片段

splitter.export_segments( output_dir="./chapters/", prefix="chapter", # chapter_01.mp3, chapter_02.mp3 format="mp3", bitrate=192 )

splitter.export_segments( output_dir="./chapters/", prefix="chapter", # 输出文件名示例：chapter_01.mp3, chapter_02.mp3 format="mp3", bitrate=192 )

Export specific segments

导出指定片段

splitter.export_segment(0, "intro.mp3") splitter.export_segment(3, "conclusion.mp3")

Save modified audio

保存修改后的音频

splitter.save("output.mp3")

undefined

splitter.save("output.mp3")

undefined

CLI Usage

命令行界面使用方法

bash

undefined

bash

undefined

Split podcast into chapters

将播客分割为章节

python podcast_splitter.py --input episode.mp3 --output-dir ./chapters/

Detect and list silences (no splitting)

检测并列出静音（不进行分割）

python podcast_splitter.py --input episode.mp3 --detect-only

Remove long silences

移除长静音

python podcast_splitter.py --input raw.mp3 --output clean.mp3 --remove-silence 2000

Custom sensitivity

自定义灵敏度

python podcast_splitter.py --input episode.mp3 --output-dir ./chapters/
--threshold -35 --min-silence 2000 --keep-silence 500

undefined

python podcast_splitter.py --input episode.mp3 --output-dir ./chapters/
--threshold -35 --min-silence 2000 --keep-silence 500

undefined

CLI Arguments

命令行参数

Argument	Description	Default
`--input`	Input audio file	Required
`--output`	Output file (for silence removal)	-
`--output-dir`	Output directory for segments	-
`--detect-only`	Only detect/report silences	False
`--threshold`	Silence threshold (dBFS)	-40
`--min-silence`	Minimum silence to detect (ms)	1000
`--keep-silence`	Silence to keep at edges (ms)	300
`--max-segments`	Maximum segments to create	None
`--remove-silence`	Remove silences longer than (ms)	-
`--shorten-silence`	Cap silence length at (ms)	-
`--prefix`	Output filename prefix	segment
`--format`	Output format	mp3
`--bitrate`	Output bitrate (kbps)	192

参数	描述	默认值
`--input`	输入音频文件	必填
`--output`	输出文件（用于静音移除）	-
`--output-dir`	片段输出目录	-
`--detect-only`	仅检测/报告静音	否
`--threshold`	静音阈值（dBFS）	-40
`--min-silence`	检测的最小静音时长（毫秒）	1000
`--keep-silence`	片段边缘保留的静音时长（毫秒）	300
`--max-segments`	最大片段数量	无
`--remove-silence`	移除长于该时长的静音（毫秒）	-
`--shorten-silence`	静音时长上限（毫秒）	-
`--prefix`	输出文件名前缀	segment
`--format`	输出格式	mp3
`--bitrate`	输出比特率（kbps）	192

Examples

示例

Split Interview into Q&A Segments

将访谈分割为问答片段

python

splitter = PodcastSplitter(
    "interview.mp3",
    silence_thresh=-35,     # Less sensitive (louder threshold)
    min_silence_len=2000,   # Only split on 2+ second pauses
    keep_silence=400        # Keep some silence for natural feel
)

segments = splitter.split_by_silence()
print(f"Found {len(segments)} segments")

splitter.export_segments("./questions/", prefix="qa")

python

splitter = PodcastSplitter(
    "interview.mp3",
    silence_thresh=-35,     # 灵敏度较低（阈值更高）
    min_silence_len=2000,   # 仅在静音时长≥2秒处分割
    keep_silence=400        # 保留少量静音以保证自然感
)

segments = splitter.split_by_silence()
print(f"检测到 {len(segments)} 个片段")

splitter.export_segments("./questions/", prefix="qa")

Remove Dead Air from Recording

从录音中移除空白静音

python

splitter = PodcastSplitter("raw_recording.mp3")

python

splitter = PodcastSplitter("raw_recording.mp3")

Show what would be removed

查看将被移除的静音

splitter.print_silence_report()

Remove silences longer than 3 seconds

移除时长超过3秒的静音

splitter.remove_silence(min_length=3000)

Cap remaining silences at 1 second

将剩余静音时长限制为1秒

splitter.shorten_silence(max_length=1000)

splitter.save("clean_recording.mp3")

undefined

splitter.shorten_silence(max_length=1000)

splitter.save("clean_recording.mp3")

undefined

Create Highlight Clips

创建精彩片段

python

splitter = PodcastSplitter("episode.mp3")
segments = splitter.split_by_silence(min_silence_len=5000)

python

splitter = PodcastSplitter("episode.mp3")
segments = splitter.split_by_silence(min_silence_len=5000)

Export only segments longer than 30 seconds

仅导出时长超过30秒的片段

for i, segment in enumerate(segments): duration = segment['end'] - segment['start'] if duration > 30000: # > 30 seconds splitter.export_segment(i, f"highlight_{i+1}.mp3")

undefined

for i, segment in enumerate(segments): duration = segment['end'] - segment['start'] if duration > 30000: # > 30秒 splitter.export_segment(i, f"highlight_{i+1}.mp3")

undefined

Batch Process Episodes

批量处理播客剧集

python

import os
from scripts.podcast_splitter import PodcastSplitter

episodes_dir = "./raw_episodes/"
output_dir = "./processed/"

for filename in os.listdir(episodes_dir):
    if filename.endswith('.mp3'):
        filepath = os.path.join(episodes_dir, filename)
        splitter = PodcastSplitter(filepath)

        # Remove long silences
        splitter.remove_silence(min_length=2000)

        # Save cleaned version
        output_path = os.path.join(output_dir, filename)
        splitter.save(output_path)
        print(f"Processed: {filename}")

python

import os
from scripts.podcast_splitter import PodcastSplitter

episodes_dir = "./raw_episodes/"
output_dir = "./processed/"

for filename in os.listdir(episodes_dir):
    if filename.endswith('.mp3'):
        filepath = os.path.join(episodes_dir, filename)
        splitter = PodcastSplitter(filepath)

        # 移除长静音
        splitter.remove_silence(min_length=2000)

        # 保存处理后的版本
        output_path = os.path.join(output_dir, filename)
        splitter.save(output_path)
        print(f"已处理：{filename}")

Preview Silence Detection

预览静音检测结果

python

splitter = PodcastSplitter("episode.mp3")

python

splitter = PodcastSplitter("episode.mp3")

Get detailed silence info

获取详细静音信息

silences = splitter.detect_silence()

print("Detected Silences:") for i, (start, end) in enumerate(silences): duration = (end - start) / 1000 start_time = start / 1000 print(f" {i+1}. {start_time:.1f}s - {duration:.1f}s silence")

silences = splitter.detect_silence()

print("检测到的静音：") for i, (start, end) in enumerate(silences): duration = (end - start) / 1000 start_time = start / 1000 print(f" {i+1}. {start_time:.1f}秒 - 静音时长 {duration:.1f}秒")

Print summary

打印摘要

splitter.print_silence_report()

undefined

splitter.print_silence_report()

undefined

Detection Settings Guide

检测设置指南

Audio Type	Threshold	Min Silence	Notes
Quiet studio	-50 dBFS	500ms	Very sensitive
Normal podcast	-40 dBFS	1000ms	Default
Noisy recording	-35 dBFS	1500ms	Less sensitive
Music with breaks	-30 dBFS	2000ms	For spoken breaks

音频类型	阈值	最小静音时长	说明
安静录音室	-50 dBFS	500毫秒	灵敏度高
普通播客	-40 dBFS	1000毫秒	默认设置
噪音录音	-35 dBFS	1500毫秒	灵敏度低
带停顿的音乐	-30 dBFS	2000毫秒	适用于语音停顿场景

Adjusting Sensitivity

调整灵敏度

More splits: Lower threshold (e.g., -50), shorter min_silence
Fewer splits: Higher threshold (e.g., -30), longer min_silence
Natural feel: Longer keep_silence (500-1000ms)
Tight edits: Shorter keep_silence (100-200ms)

更多片段：降低阈值（如-50），缩短最小静音时长
更少片段：提高阈值（如-30），延长最小静音时长
自然效果：延长边缘保留静音时长（500-1000毫秒）
紧凑编辑：缩短边缘保留静音时长（100-200毫秒）

Dependencies

依赖项

pydub>=0.25.0

Note: Requires FFmpeg installed on system.

pydub>=0.25.0

注意：需要在系统上安装FFmpeg。

Limitations

局限性

Works best with speech content (not music)
Very noisy recordings may need threshold adjustment
Long files use significant memory
No automatic chapter naming (manual rename needed)

对语音内容效果最佳（不适用于音乐）
噪音较大的录音可能需要调整阈值
长文件会占用大量内存
不支持自动章节命名（需要手动重命名）