工作流 /
音频 /
Chatter Box：语音克隆音频转换
Chatter Box：语音克隆音频转换
Chatter Box: Voice Cloning Audio Conversion
上传音频片段和简短语音参考，生成转换后的音频输出，克隆目标声音同时保留原始节奏和表现。
音频
6 个节点 2 个输入 21 次使用 ComfyUI ≥ 0.25.0 PurzBeats 6 KB
节点构成
LoadAudio
× 2
AudioCrop
× 1
FL_ChatterboxVC
× 1
MarkdownNote
× 1
SaveAudioMP3
× 1
官方标注的模型
Chatter Box
工作流 JSON
和上面「下载工作流 JSON」是同一份文件。
1{
2 "id": "18bb6f1f-5453-484d-9a6d-260b40f4e887",
3 "revision": 0,
4 "last_node_id": 7,
5 "last_link_id": 11,
6 "nodes": [
7 {
8 "id": 2,
9 "type": "LoadAudio",
10 "pos": [
11 -1062.0764835957827,
12 359.223383951974
13 ],
14 "size": [
15 282.798828125,
16 136
17 ],
18 "flags": [],
19 "order": 0,
20 "mode": 0,
21 "inputs": [],
22 "outputs": [
23 {
24 "name": "AUDIO",
25 "type": "AUDIO",
26 "links": [
27 11
28 ]
29 }
30 ],
31 "properties": {
32 "Node name for S&R": "LoadAudio"
33 },
34 "widgets_values": [
35 "chatterbox_input_target_voice.mp3",
36 null,
37 null
38 ]
39 },
40 {
41 "id": 5,
42 "type": "AudioCrop",
43 "pos": [
44 -738.5534728009812,
45 170.18822419548158
46 ],
47 "size": [
48 270,
49 82
50 ],
51 "flags": [],
52 "order": 3,
53 "mode": 0,
54 "inputs": [
55 {
56 "name": "audio",
57 "type": "AUDIO",
58 "link": 5
59 }
60 ],
61 "outputs": [
62 {
63 "name": "AUDIO",
64 "type": "AUDIO",
65 "links": [
66 10
67 ]
68 }
69 ],
70 "properties": {
71 "Node name for S&R": "AudioCrop"
72 },
73 "widgets_values": [
74 "0:00",
75 "0:27"
76 ]
77 },
78 {
79 "id": 1,
80 "type": "FL_ChatterboxVC",
81 "pos": [
82 -427.25661449685225,
83 167.3668274611146
84 ],
85 "size": [
86 277.015625,
87 150
88 ],
89 "flags": [],
90 "order": 4,
91 "mode": 0,
92 "inputs": [
93 {
94 "name": "input_audio",
95 "type": "AUDIO",
96 "link": 10
97 },
98 {
99 "name": "target_voice",
100 "type": "AUDIO",
101 "link": 11
102 }
103 ],
104 "outputs": [
105 {
106 "name": "audio",
107 "type": "AUDIO",
108 "links": [
109 4
110 ]
111 },
112 {
113 "name": "message",
114 "type": "STRING",
115 "links": null
116 }
117 ],
118 "properties": {
119 "Node name for S&R": "FL_ChatterboxVC"
120 },
121 "widgets_values": [
122 1956129161,
123 "randomize",
124 false,
125 false
126 ]
127 },
128 {
129 "id": 7,
130 "type": "MarkdownNote",
131 "pos": [
132 -1500,
133 170
134 ],
135 "size": [
136 402.81050958333503,
137 667.0051882662111
138 ],
139 "flags": [],
140 "order": 1,
141 "mode": 0,
142 "inputs": [],
143 "outputs": [],
144 "title": "🎙️ Chatterbox Voice Cloning (Audio → Audio)",
145 "properties": [],
146 "widgets_values": [
147 "This workflow takes one audio clip and converts it into another voice while preserving the original timing and performance.\n\n---\n\n## What This Does\n\n- **Input audio**: the speech or performance you want to convert \n- **Target voice**: a short reference clip of the voice you want to clone \n- **Output**: the original audio re-spoken in the target voice\n\n---\n\n## How to Use\n\n### 1. Load the Source Audio\n- Upload the audio you want to convert into the **top Load Audio** node \n- This is the performance that will be re-voiced\n\n### 2. (Optional) Trim the Audio\n- Use **AudioCrop** to select a specific section \n- Adjust `start_time` and `end_time` to isolate the part you want\n- You can bypass the **AudioCrop** node if you don't need to trim your clip\n\n### 3. Load the Target Voice\n- Upload a short reference clip into the **bottom Load Audio** node \n- 5–10 seconds of clean speech works well \n- This audio defines the cloned voice\n\n### 4. Generate the Voice Conversion\n- The **FL Chatterbox VC** node takes:\n - `input_audio` → the source performance\n - `target_voice` → the voice reference\n- Click **Run** to generate\n\n### 5. Listen or Export\n- Preview the result in **Save Audio (MP3)** \n- The output keeps the original timing, phrasing, and cadence\n\n---\n\n## Tips\n\n- Clean, single-speaker reference clips give the best results \n- Shorter, well-cropped segments are more reliable \n- If results sound off, try trimming silence or reducing background noise\n\n---\n"
148 ],
149 "color": "#222",
150 "bgcolor": "#000"
151 },
152 {
153 "id": 4,
154 "type": "SaveAudioMP3",
155 "pos": [
156 -116.61273975353727,
157 165.40033076921117
158 ],
159 "size": [
160 270,
161 136
162 ],
163 "flags": [],
164 "order": 5,
165 "mode": 0,
166 "inputs": [
167 {
168 "name": "audio",
169 "type": "AUDIO",
170 "link": 4
171 }
172 ],
173 "outputs": [],
174 "properties": {
175 "Node name for S&R": "SaveAudioMP3"
176 },
177 "widgets_values": [
178 "audio/ComfyUI",
179 "V0"
180 ]
181 },
182 {
183 "id": 3,
184 "type": "LoadAudio",
185 "pos": [
186 -1063.0168912935958,
187 168.30728251554427
188 ],
189 "size": [
190 282.798828125,
191 136
192 ],
193 "flags": [],
194 "order": 2,
195 "mode": 0,
196 "inputs": [],
197 "outputs": [
198 {
199 "name": "AUDIO",
200 "type": "AUDIO",
201 "links": [
202 5
203 ]
204 }
205 ],
206 "properties": {
207 "Node name for S&R": "LoadAudio"
208 },
209 "widgets_values": [
210 "chatterbox_input_gettysburg.wav",
211 null,
212 null
213 ]
214 }
215 ],
216 "links": [
217 [
218 4,
219 1,
220 0,
221 4,
222 0,
223 "AUDIO"
224 ],
225 [
226 5,
227 3,
228 0,
229 5,
230 0,
231 "AUDIO"
232 ],
233 [
234 10,
235 5,
236 0,
237 1,
238 0,
239 "AUDIO"
240 ],
241 [
242 11,
243 2,
244 0,
245 1,
246 1,
247 "AUDIO"
248 ]
249 ],
250 "groups": [],
251 "config": [],
252 "extra": {
253 "workflowRendererVersion": "LG",
254 "ds": {
255 "scale": 0.5875498691146271,
256 "offset": [
257 2023.9183970432728,
258 500.497443710502
259 ]
260 },
261 "frontendVersion": "1.37.10",
262 "VHS_latentpreview": false,
263 "VHS_latentpreviewrate": 0,
264 "VHS_MetadataImage": true,
265 "VHS_KeepIntermediate": true
266 },
267 "version": 0.4
268}