工作流 /
音频 /
Breeze TTS 2:文本转语音
Breeze TTS 2:文本转语音
Breeze TTS 2: Text to Speech
使用 Breeze TTS 2 把文字变成语音——这是 Breeze Blue 的开源权重语音模型,原生支持中英文,并可用多语言版本覆盖 51 种语言。用三秒录音克隆一个声音,或用文字描述设计全新声音,也可以直接选用内置声音,再用自然语言指令或 (sighs) 这类内联音频标签指导表演。工作流自带示例录音、可复用的 voice id 保存链路,以及可选的 Remix 与 Design 工具。适合旁白、角色对白和多语种配音。
音频
17 个节点 1 个输入 ComfyUI ≥ 0.45.4 ComfyUI 25 KB
节点构成
MarkdownNote
× 5
BreezeVoiceSelector
× 2
BreezeTextToSpeech
× 1
SaveAudioAdvanced
× 1
BreezeVoiceClone
× 1
PreviewAny
× 1
LoadAudio
× 1
SaveText
× 1
BreezeVoiceRemix
× 1
BreezeVoiceDesign
× 1
StringConcatenate
× 1
RecordAudio
× 1
官方标注的模型
Breeze Blue
工作流 JSON
和上面「下载工作流 JSON」是同一份文件。
1{
2 "id": "dea2ff39-c2ce-4f22-8a30-75ecdc6c59b2",
3 "revision": 0,
4 "last_node_id": 29,
5 "last_link_id": 14,
6 "nodes": [
7 {
8 "id": 1,
9 "type": "BreezeTextToSpeech",
10 "pos": [
11 1990,
12 -1680
13 ],
14 "size": [
15 400,
16 260
17 ],
18 "flags": [],
19 "order": 14,
20 "mode": 0,
21 "inputs": [
22 {
23 "name": "voice",
24 "type": "BREEZE_VOICE",
25 "link": 14
26 }
27 ],
28 "outputs": [
29 {
30 "name": "AUDIO",
31 "type": "AUDIO",
32 "links": [
33 1
34 ]
35 }
36 ],
37 "properties": {
38 "Node name for S&R": "BreezeTextToSpeech"
39 },
40 "widgets_values": [
41 "Meet the Breeze TTS. (sighs) It weighs less than a paperback and folds flat in your bag. Titanium frame, recycled shell, and a hinge that still feels new after ten thousand openings. Available Thursday, wherever good gear is sold.",
42 "English",
43 "Breeze TTS 2",
44 "neutral",
45 false,
46 1,
47 767741007,
48 "randomize"
49 ],
50 "widgets_values_named": {
51 "text": "Meet the Breeze TTS. (sighs) It weighs less than a paperback and folds flat in your bag. Titanium frame, recycled shell, and a hinge that still feels new after ten thousand openings. Available Thursday, wherever good gear is sold.",
52 "language": "English",
53 "language.model": "Breeze TTS 2",
54 "delivery": "neutral",
55 "delivery.speaker_boost": false,
56 "speed": 1,
57 "seed": 767741007,
58 "control_after_generate": "randomize"
59 },
60 "color": "#432",
61 "bgcolor": "#653"
62 },
63 {
64 "id": 2,
65 "type": "SaveAudioAdvanced",
66 "pos": [
67 2580,
68 -1680
69 ],
70 "size": [
71 450,
72 270
73 ],
74 "flags": [],
75 "order": 16,
76 "mode": 0,
77 "inputs": [
78 {
79 "name": "audio",
80 "type": "AUDIO",
81 "link": 1
82 }
83 ],
84 "outputs": [
85 {
86 "name": "audio",
87 "type": "AUDIO",
88 "links": null
89 }
90 ],
91 "properties": [],
92 "widgets_values": [
93 "audio/Breeze_TTS2",
94 "flac"
95 ],
96 "widgets_values_named": {
97 "filename_prefix": "audio/Breeze_TTS2",
98 "format": "flac"
99 }
100 },
101 {
102 "id": 4,
103 "type": "BreezeVoiceClone",
104 "pos": [
105 340,
106 -1660
107 ],
108 "size": [
109 350,
110 140
111 ],
112 "flags": [],
113 "order": 10,
114 "mode": 0,
115 "inputs": [
116 {
117 "name": "audio",
118 "type": "AUDIO",
119 "link": 4
120 }
121 ],
122 "outputs": [
123 {
124 "name": "voice",
125 "type": "BREEZE_VOICE",
126 "links": [
127 6
128 ]
129 },
130 {
131 "name": "preview",
132 "type": "AUDIO",
133 "links": null
134 }
135 ],
136 "properties": {
137 "Node name for S&R": "BreezeVoiceClone"
138 },
139 "widgets_values": [
140 false,
141 892527809,
142 "randomize"
143 ],
144 "widgets_values_named": {
145 "remove_background_noise": false,
146 "seed": 892527809,
147 "control_after_generate": "randomize"
148 },
149 "color": "#432",
150 "bgcolor": "#653"
151 },
152 {
153 "id": 8,
154 "type": "PreviewAny",
155 "pos": [
156 750,
157 -1660
158 ],
159 "size": [
160 210,
161 130
162 ],
163 "flags": [],
164 "order": 11,
165 "mode": 0,
166 "inputs": [
167 {
168 "name": "source",
169 "type": "*",
170 "link": 6
171 }
172 ],
173 "outputs": [
174 {
175 "name": "STRING",
176 "type": "STRING",
177 "links": [
178 7,
179 12
180 ]
181 }
182 ],
183 "title": "Preview as Text (voice id)",
184 "properties": {
185 "Node name for S&R": "PreviewAny"
186 },
187 "widgets_values": [],
188 "widgets_values_named": []
189 },
190 {
191 "id": 9,
192 "type": "LoadAudio",
193 "pos": [
194 -130,
195 -1660
196 ],
197 "size": [
198 400,
199 170
200 ],
201 "flags": [],
202 "order": 0,
203 "mode": 0,
204 "inputs": [],
205 "outputs": [
206 {
207 "name": "AUDIO",
208 "type": "AUDIO",
209 "links": [
210 4
211 ]
212 }
213 ],
214 "properties": {
215 "Node name for S&R": "LoadAudio"
216 },
217 "widgets_values": [
218 "breeze_tts2_tts_clone_audio.wav",
219 ""
220 ],
221 "widgets_values_named": {
222 "audio": "clone_tts.wav",
223 "upload": ""
224 }
225 },
226 {
227 "id": 11,
228 "type": "BreezeVoiceSelector",
229 "pos": [
230 1020,
231 -1660
232 ],
233 "size": [
234 270,
235 90
236 ],
237 "flags": [],
238 "order": 12,
239 "mode": 0,
240 "inputs": [
241 {
242 "name": "language.voice_id",
243 "type": "STRING",
244 "widget": {
245 "name": "language.voice_id"
246 },
247 "link": 7
248 }
249 ],
250 "outputs": [
251 {
252 "name": "voice",
253 "type": "BREEZE_VOICE",
254 "links": [
255 14
256 ]
257 }
258 ],
259 "properties": {
260 "Node name for S&R": "BreezeVoiceSelector"
261 },
262 "widgets_values": [
263 "custom",
264 ""
265 ],
266 "widgets_values_named": {
267 "language": "custom",
268 "language.voice_id": ""
269 }
270 },
271 {
272 "id": 13,
273 "type": "SaveText",
274 "pos": [
275 1120,
276 -1400
277 ],
278 "size": [
279 280,
280 230
281 ],
282 "flags": [],
283 "order": 15,
284 "mode": 0,
285 "inputs": [
286 {
287 "name": "text",
288 "type": "STRING",
289 "link": 13
290 }
291 ],
292 "outputs": [
293 {
294 "name": "text",
295 "type": "STRING",
296 "links": null
297 }
298 ],
299 "properties": {
300 "Node name for S&R": "SaveText"
301 },
302 "widgets_values": [
303 "Breeze_voice_id",
304 "txt"
305 ],
306 "widgets_values_named": {
307 "filename_prefix": "Breeze_voice_id",
308 "format": "txt"
309 }
310 },
311 {
312 "id": 15,
313 "type": "BreezeVoiceSelector",
314 "pos": [
315 1590,
316 -1540
317 ],
318 "size": [
319 270,
320 90
321 ],
322 "flags": [],
323 "order": 1,
324 "mode": 0,
325 "inputs": [],
326 "outputs": [
327 {
328 "name": "voice",
329 "type": "BREEZE_VOICE",
330 "links": []
331 }
332 ],
333 "properties": {
334 "Node name for S&R": "BreezeVoiceSelector"
335 },
336 "widgets_values": [
337 "English",
338 "Elizabeth"
339 ],
340 "widgets_values_named": {
341 "language": "English",
342 "language.voice": "Elizabeth"
343 }
344 },
345 {
346 "id": 17,
347 "type": "BreezeVoiceRemix",
348 "pos": [
349 -130,
350 -910
351 ],
352 "size": [
353 400,
354 204
355 ],
356 "flags": [],
357 "order": 2,
358 "mode": 4,
359 "inputs": [
360 {
361 "name": "voice",
362 "type": "BREEZE_VOICE",
363 "link": null
364 }
365 ],
366 "outputs": [
367 {
368 "name": "voice",
369 "type": "BREEZE_VOICE",
370 "links": null
371 },
372 {
373 "name": "preview",
374 "type": "AUDIO",
375 "links": null
376 }
377 ],
378 "properties": {
379 "Node name for S&R": "BreezeVoiceRemix"
380 },
381 "widgets_values": [
382 "custom",
383 "",
384 4,
385 1201060479,
386 "randomize"
387 ],
388 "widgets_values_named": {
389 "change": "custom",
390 "change.prompt": "",
391 "strength": 4,
392 "seed": 1201060479,
393 "control_after_generate": "randomize"
394 },
395 "color": "#432",
396 "bgcolor": "#653"
397 },
398 {
399 "id": 18,
400 "type": "BreezeVoiceDesign",
401 "pos": [
402 666,
403 -910
404 ],
405 "size": [
406 420,
407 325.912524242866
408 ],
409 "flags": [],
410 "order": 3,
411 "mode": 4,
412 "inputs": [],
413 "outputs": [
414 {
415 "name": "voice",
416 "type": "BREEZE_VOICE",
417 "links": null
418 },
419 {
420 "name": "preview",
421 "type": "AUDIO",
422 "links": null
423 }
424 ],
425 "properties": {
426 "Node name for S&R": "BreezeVoiceDesign"
427 },
428 "widgets_values": [
429 "",
430 "",
431 "English",
432 "Breeze TTS 2",
433 4,
434 920438200,
435 "randomize"
436 ],
437 "widgets_values_named": {
438 "description": "",
439 "preview_text": "",
440 "language": "English",
441 "language.model": "Breeze TTS 2",
442 "strength": 4,
443 "seed": 920438200,
444 "control_after_generate": "randomize"
445 },
446 "color": "#432",
447 "bgcolor": "#653"
448 },
449 {
450 "id": 20,
451 "type": "MarkdownNote",
452 "pos": [
453 317.18685378005824,
454 -1411.8023414402815
455 ],
456 "size": [
457 400,
458 245
459 ],
460 "flags": [],
461 "order": 4,
462 "mode": 0,
463 "inputs": [],
464 "outputs": [],
465 "title": "Note: clone → saved voice id",
466 "properties": [],
467 "widgets_values": [
468 "**Voice clone — input to reusable voice id**\n\n`Load Audio → Breeze Voice Clone → Preview as Text`\n\n- One speaker, **at least 3 s**. Only the first **60 s** of the file are read;\n Breeze keeps the best ≤30 s, trims it clean and normalises the loudness.\n- `remove_background_noise` helps on a rough phone take.\n- `voice` output = the new **voice id** (`voc_…`), shown by *Preview as Text*.\n `preview` output = how it sounds.\n- Only clone a voice you have the right to use — consent is required.\n\n**Reuse the id later:** paste it into *Breeze Voice Selector → custom*.\n*Concatenate Text* adds a readable prefix and *Save Text* writes the id to a\n`.txt` for your records.\n\nSample to try: [breeze_tts2_tts_clone_audio.wav](https://raw.githubusercontent.com/Comfy-Org/workflow_templates/refs/heads/main/input/breeze_tts2_tts_clone_audio.wav)."
469 ],
470 "widgets_values_named": {
471 "text": "**Voice clone — input to reusable voice id**\n\n`Load Audio → Breeze Voice Clone → Preview as Text`\n\n- One speaker, **at least 3 s**. Only the first **60 s** of the file are read;\n Breeze keeps the best ≤30 s, trims it clean and normalises the loudness.\n- `remove_background_noise` helps on a rough phone take.\n- `voice` output = the new **voice id** (`voc_…`), shown by *Preview as Text*.\n `preview` output = how it sounds.\n- Only clone a voice you have the right to use — consent is required.\n\n**Reuse the id later:** paste it into *Breeze Voice Selector → custom*.\n*Concatenate Text* adds a readable prefix and *Save Text* writes the id to a\n`.txt` for your records.\n"
472 },
473 "color": "#222",
474 "bgcolor": "#000"
475 },
476 {
477 "id": 21,
478 "type": "StringConcatenate",
479 "pos": [
480 760,
481 -1400
482 ],
483 "size": [
484 300,
485 190
486 ],
487 "flags": [],
488 "order": 13,
489 "mode": 0,
490 "inputs": [
491 {
492 "name": "string_b",
493 "type": "STRING",
494 "widget": {
495 "name": "string_b"
496 },
497 "link": 12
498 }
499 ],
500 "outputs": [
501 {
502 "name": "STRING",
503 "type": "STRING",
504 "links": [
505 13
506 ]
507 }
508 ],
509 "title": "Concatenate Text(append prefix for id)",
510 "properties": {
511 "Node name for S&R": "StringConcatenate"
512 },
513 "widgets_values": [
514 "you custom id name here: ",
515 "",
516 ""
517 ],
518 "widgets_values_named": {
519 "string_a": "you custom id name here: ",
520 "string_b": "",
521 "delimiter": ""
522 }
523 },
524 {
525 "id": 24,
526 "type": "MarkdownNote",
527 "pos": [
528 303,
529 -910
530 ],
531 "size": [
532 322.9329231665266,
533 383.0390404970097
534 ],
535 "flags": [],
536 "order": 5,
537 "mode": 0,
538 "inputs": [],
539 "outputs": [],
540 "title": "Note: Voice Remix",
541 "properties": [],
542 "widgets_values": [
543 "## Breeze Voice Remix\n\nMakes a **changed copy** of a voice you already have, in that voice's language.\n\n- `change = custom` — describe the shift in words: \"older, huskier, tired, a\n little slower\".\n- `change = preset` — 27 named shifts:\n **age** (teenage → elderly) · **gender** (male imitating female, and reverse) ·\n **pitch** (higher / lower) · **pace** (faster / slower) ·\n **energy** (energetic / low-energy) ·\n **emotion** (excited, angry, sad, afraid, awkward, sarcastic, restrained\n grief, painful) ·\n **role-play** (stage actor, news anchor, bedtime storyteller) ·\n **delivery** (laughing, crying, whispered, singing).\n- `strength` 1–10 — how far it moves from the original. 4 is a good start.\n\nOutputs: a new **voice id** (saved to your account) and an audio **preview**.\nSend the voice to *Breeze Text to Speech*, or save the id for later runs.\n\nNeed another language? Use *Breeze Voice Localize* — Remix stays in the same\nlanguage.\n"
544 ],
545 "widgets_values_named": {
546 "text": "## Breeze Voice Remix\n\nMakes a **changed copy** of a voice you already have, in that voice's language.\n\n- `change = custom` — describe the shift in words: \"older, huskier, tired, a\n little slower\".\n- `change = preset` — 27 named shifts:\n **age** (teenage → elderly) · **gender** (male imitating female, and reverse) ·\n **pitch** (higher / lower) · **pace** (faster / slower) ·\n **energy** (energetic / low-energy) ·\n **emotion** (excited, angry, sad, afraid, awkward, sarcastic, restrained\n grief, painful) ·\n **role-play** (stage actor, news anchor, bedtime storyteller) ·\n **delivery** (laughing, crying, whispered, singing).\n- `strength` 1–10 — how far it moves from the original. 4 is a good start.\n\nOutputs: a new **voice id** (saved to your account) and an audio **preview**.\nSend the voice to *Breeze Text to Speech*, or save the id for later runs.\n\nNeed another language? Use *Breeze Voice Localize* — Remix stays in the same\nlanguage.\n"
547 },
548 "color": "#222",
549 "bgcolor": "#000"
550 },
551 {
552 "id": 25,
553 "type": "MarkdownNote",
554 "pos": [
555 1119,
556 -910
557 ],
558 "size": [
559 308.01025331063624,
560 426.03903633031507
561 ],
562 "flags": [],
563 "order": 6,
564 "mode": 0,
565 "inputs": [],
566 "outputs": [],
567 "title": "Note: Voice Design",
568 "properties": [],
569 "widgets_values": [
570 "## Breeze Voice Design\n\nInvents a **new voice from words** — no reference audio needed.\n\n- `description` (3–500 chars) — who the voice is: age, gender, accent, timbre,\n pace, character. e.g. \"Warm male voice in his mid-forties with a soft Scottish\n accent, close-miked, unhurried, with dry humour underneath.\"\n- `preview_text` (≤500 chars) — the line the new voice speaks. **The voice is\n built from this line**, so write it in the target language and in the tone you\n want; audio tags such as `(sighs)` work here too.\n- `language` + `model` — same rule as text to speech: English and Chinese can use\n either model, every other language uses the multilingual one.\n- `strength` 1–10 — how literally it follows the description.\n- Reroll `seed` for another take on the same description.\n\nOutputs: the **voice id** (named after the description, saved to your account)\nand an audio **preview**. Send the voice to *Breeze Text to Speech*.\n"
571 ],
572 "widgets_values_named": {
573 "text": "## Breeze Voice Design\n\nInvents a **new voice from words** — no reference audio needed.\n\n- `description` (3–500 chars) — who the voice is: age, gender, accent, timbre,\n pace, character. e.g. \"Warm male voice in his mid-forties with a soft Scottish\n accent, close-miked, unhurried, with dry humour underneath.\"\n- `preview_text` (≤500 chars) — the line the new voice speaks. **The voice is\n built from this line**, so write it in the target language and in the tone you\n want; audio tags such as `(sighs)` work here too.\n- `language` + `model` — same rule as text to speech: English and Chinese can use\n either model, every other language uses the multilingual one.\n- `strength` 1–10 — how literally it follows the description.\n- Reroll `seed` for another take on the same description.\n\nOutputs: the **voice id** (named after the description, saved to your account)\nand an audio **preview**. Send the voice to *Breeze Text to Speech*.\n"
574 },
575 "color": "#222",
576 "bgcolor": "#000"
577 },
578 {
579 "id": 27,
580 "type": "MarkdownNote",
581 "pos": [
582 -813.1042406846639,
583 -1751.232953656752
584 ],
585 "size": [
586 600,
587 770
588 ],
589 "flags": [],
590 "order": 7,
591 "mode": 0,
592 "inputs": [],
593 "outputs": [],
594 "title": "Note: Breeze TTS 2 — read me first",
595 "properties": [],
596 "widgets_values": [
597 "# Breeze TTS 2 — voice toolkit\n\n**Breeze TTS 2** (by Breeze Blue) is an open-weight text-to-speech model built for\nreal-time speech. One model speaks **English and Chinese** natively and a\nmultilingual variant covers **51 languages**. Delivery is steerable with plain\nwords, and a voice can be cloned from a 3-second sample or designed from a\nwritten description.\n\nThese are **partner nodes**: they run on Comfy's servers, so sign in to Comfy and\nkeep API credits available.\n\n| Model | Languages |\n| --- | --- |\n| Breeze TTS 2 | English, Chinese |\n| Breeze TTS 2 Multilingual | all 51 (only option outside English/Chinese) |\n\n## What this workflow does\n\n**Step 1 — create a voice**\n`Load Audio → Breeze Voice Clone → Preview as Text`\nThe clone node returns a **voice id** (`voc_…`) plus a spoken preview. The id is\nshown in *Preview as Text*, gets a readable prefix, and *Save Text* writes it to\na `.txt` so you can reuse it in later runs.\n\n**Step 2 — speak the script**\n`Breeze Voice Selector (custom) → Breeze Text to Speech → Save Audio`\nThe selector passes the cloned id into *Breeze Text to Speech*. Output lands in\n`ComfyUI/output/audio/Breeze_TTS2…`.\n\n**Optional — a featured voice** (group on the right)\n*Breeze Voice Selector* set to `English → Elizabeth` is a ready-made voice: bypass\nthe clone branch and wire it where the custom selector sits.\n\n**Voice tools** (bottom group — select all, then **Ctrl+B** to enable)\n*Breeze Voice Remix* reshapes a voice you already have; *Breeze Voice Design*\ninvents a new one from words. Each returns a voice id + preview; connect either\nto *Breeze Text to Speech*.\n\n## Writing the script\n\n- 1–1000 characters per request.\n- Audio tags are **performed, not read**: `(sighs)`, `(laughs)`, `(whispers)` in\n English — `[叹气]`, `[笑]`, `[咳嗽]` in Chinese.\n- `delivery = neutral` keeps the voice as it is; `speaker_boost` pulls it closer\n to the original sound. `delivery = directed` follows `instructions` written in\n the text's language, and `strength` (1–10) sets how hard it leans on them.\n- `speed` 0.5–2.0 changes tempo, never pitch.\n- A voice sounds best in its own language. For another language, make a native\n copy with *Breeze Voice Localize* instead of feeding foreign text to the voice.\n\n## Input asset\n\n- [breeze_tts2_tts_clone_audio.wav](https://raw.githubusercontent.com/Comfy-Org/workflow_templates/refs/heads/main/input/breeze_tts2_tts_clone_audio.wav) — the sample recording wired into `Load Audio`. Download it into `ComfyUI/input/`, or swap `Load Audio` for **Record Audio** and capture a take straight from the mic.\n- Any clean single-speaker clip of 3 s or more works.\n"
598 ],
599 "widgets_values_named": {
600 "text": "# Breeze TTS 2 — voice toolkit\n\n**Breeze TTS 2** (by Breeze Blue) is an open-weight text-to-speech model built for\nreal-time speech. One model speaks **English and Chinese** natively and a\nmultilingual variant covers **51 languages**. Delivery is steerable with plain\nwords, and a voice can be cloned from a 3-second sample or designed from a\nwritten description.\n\nThese are **partner nodes**: they run on Comfy's servers, so sign in to Comfy and\nkeep API credits available.\n\n| Model | Languages |\n| --- | --- |\n| Breeze TTS 2 | English, Chinese |\n| Breeze TTS 2 Multilingual | all 51 (only option outside English/Chinese) |\n\n## What this workflow does\n\n**Step 1 — create a voice**\n`Load Audio → Breeze Voice Clone → Preview as Text`\nThe clone node returns a **voice id** (`voc_…`) plus a spoken preview. The id is\nshown in *Preview as Text*, gets a readable prefix, and *Save Text* writes it to\na `.txt` so you can reuse it in later runs.\n\n**Step 2 — speak the script**\n`Breeze Voice Selector (custom) → Breeze Text to Speech → Save Audio`\nThe selector passes the cloned id into *Breeze Text to Speech*. Output lands in\n`ComfyUI/output/audio/Breeze_TTS2…`.\n\n**Optional — a featured voice** (group on the right)\n*Breeze Voice Selector* set to `English → Elizabeth` is a ready-made voice: bypass\nthe clone branch and wire it where the custom selector sits.\n\n**Voice tools** (bottom group — select all, then **Ctrl+B** to enable)\n*Breeze Voice Remix* reshapes a voice you already have; *Breeze Voice Design*\ninvents a new one from words. Each returns a voice id + preview; connect either\nto *Breeze Text to Speech*.\n\n## Writing the script\n\n- 1–1000 characters per request.\n- Audio tags are **performed, not read**: `(sighs)`, `(laughs)`, `(whispers)` in\n English — `[叹气]`, `[笑]`, `[咳嗽]` in Chinese.\n- `delivery = neutral` keeps the voice as it is; `speaker_boost` pulls it closer\n to the original sound. `delivery = directed` follows `instructions` written in\n the text's language, and `strength` (1–10) sets how hard it leans on them.\n- `speed` 0.5–2.0 changes tempo, never pitch.\n- A voice sounds best in its own language. For another language, make a native\n copy with *Breeze Voice Localize* instead of feeding foreign text to the voice.\n"
601 },
602 "color": "#222",
603 "bgcolor": "#000"
604 },
605 {
606 "id": 28,
607 "type": "MarkdownNote",
608 "pos": [
609 1589.5414988308426,
610 -1328.0823256103233
611 ],
612 "size": [
613 280,
614 101.9625231726965
615 ],
616 "flags": [],
617 "order": 8,
618 "mode": 0,
619 "inputs": [],
620 "outputs": [],
621 "properties": [],
622 "widgets_values": [
623 "Currently ComfyUI doesn't have remote resource preview capability, so you can preview the built-in audio :("
624 ],
625 "widgets_values_named": {
626 "text": "Currently ComfyUI doesn't have remote resource preview capability, so you can preview the built-in audio :("
627 },
628 "color": "#222",
629 "bgcolor": "#000"
630 },
631 {
632 "id": 29,
633 "type": "RecordAudio",
634 "pos": [
635 -107.89514662745893,
636 -1383.0603730630514
637 ],
638 "size": [
639 353.8934329860688,
640 114.00000817471667
641 ],
642 "flags": [],
643 "order": 9,
644 "mode": 0,
645 "inputs": [],
646 "outputs": [
647 {
648 "name": "AUDIO",
649 "type": "AUDIO",
650 "links": null
651 }
652 ],
653 "properties": {
654 "Node name for S&R": "RecordAudio"
655 },
656 "widgets_values": [
657 "",
658 ""
659 ],
660 "widgets_values_named": {
661 "audio": "",
662 "audio#1": ""
663 }
664 }
665 ],
666 "links": [
667 [
668 4,
669 9,
670 0,
671 4,
672 0,
673 "AUDIO"
674 ],
675 [
676 6,
677 4,
678 0,
679 8,
680 0,
681 "BREEZE_VOICE"
682 ],
683 [
684 13,
685 21,
686 0,
687 13,
688 0,
689 "STRING"
690 ],
691 [
692 12,
693 8,
694 0,
695 21,
696 0,
697 "STRING"
698 ],
699 [
700 14,
701 11,
702 0,
703 1,
704 0,
705 "BREEZE_VOICE"
706 ],
707 [
708 1,
709 1,
710 0,
711 2,
712 0,
713 "AUDIO"
714 ],
715 [
716 7,
717 8,
718 0,
719 11,
720 0,
721 "STRING"
722 ]
723 ],
724 "groups": [
725 {
726 "id": 1,
727 "title": "Step 1 - Create a Voice Clone",
728 "bounding": [
729 -160,
730 -1780,
731 1610,
732 670
733 ],
734 "color": "#3f789e",
735 "flags": []
736 },
737 {
738 "id": 2,
739 "title": "Step 2 - Audio Generation",
740 "bounding": [
741 1950,
742 -1780,
743 480,
744 670
745 ],
746 "color": "#3f789e",
747 "flags": []
748 },
749 {
750 "id": 3,
751 "title": "Save voice id",
752 "bounding": [
753 754.7055500316109,
754 -1467.1546374882246,
755 660,
756 310
757 ],
758 "color": "#3f789e",
759 "flags": []
760 },
761 {
762 "id": 4,
763 "title": "Optional: Use preset audio",
764 "bounding": [
765 1560,
766 -1630,
767 340,
768 240
769 ],
770 "color": "#3f789e",
771 "flags": []
772 },
773 {
774 "id": 5,
775 "title": "Breeze Voice Tools (Select and Ctrl-B to enable)",
776 "bounding": [
777 -160,
778 -1070,
779 1610,
780 740
781 ],
782 "color": "#444",
783 "flags": []
784 },
785 {
786 "id": 6,
787 "title": "Use your voice",
788 "bounding": [
789 -117.89514662745893,
790 -1453.0603730630514,
791 373.8934329860688,
792 194.00000817471664
793 ],
794 "color": "#3f789e",
795 "flags": []
796 }
797 ],
798 "config": [],
799 "extra": {
800 "ds": {
801 "scale": 0.7201362045677494,
802 "offset": [
803 875.0681214225718,
804 1976.9852008590194
805 ]
806 },
807 "frontendVersion": "1.59.1"
808 },
809 "version": 0.4
810}