工作流 /
音频 /
Fish Audio：语音转文字
Fish Audio：语音转文字
Fish Audio: Speech to Text
使用Fish Audio ASR将音频转录为文本。系统会自动检测80种以上语言，并可选择返回词级时间戳作为JSON片段。可用于创建字幕、转录录音或从音频片段生成TextGenerate。已预加载示例音频，您可以立即测试。
音频
5 个节点 1 个输入 9 次使用 ComfyUI ≥ 0.34.0 ComfyUI 6 KB
节点构成
SaveText
× 2
LoadAudio
× 1
FishAudioSpeechToText
× 1
MarkdownNote
× 1
官方标注的模型
Fish Audio
工作流 JSON
和上面「下载工作流 JSON」是同一份文件。
1{
2 "id": "1c3621ae-965c-43f9-ba79-dc0b92fcd468",
3 "revision": 0,
4 "last_node_id": 19,
5 "last_link_id": 13,
6 "nodes": [
7 {
8 "id": 11,
9 "type": "LoadAudio",
10 "pos": [
11 -620,
12 260
13 ],
14 "size": [
15 330,
16 170
17 ],
18 "flags": [],
19 "order": 0,
20 "mode": 0,
21 "inputs": [],
22 "outputs": [
23 {
24 "name": "AUDIO",
25 "type": "AUDIO",
26 "links": [
27 10
28 ]
29 }
30 ],
31 "properties": {
32 "Node name for S&R": "LoadAudio"
33 },
34 "widgets_values": [
35 "fish_audio_example.mp3",
36 null,
37 null
38 ],
39 "widgets_values_named": {
40 "audio": "62d0a069159eda8931ad7a6bf2a867e2695cfeb088f6a6892ceb86744a516cd3.mp3",
41 "upload": null
42 }
43 },
44 {
45 "id": 13,
46 "type": "FishAudioSpeechToText",
47 "pos": [
48 -240,
49 260
50 ],
51 "size": [
52 280,
53 230
54 ],
55 "flags": [],
56 "order": 2,
57 "mode": 0,
58 "inputs": [
59 {
60 "name": "audio",
61 "type": "AUDIO",
62 "link": 10
63 }
64 ],
65 "outputs": [
66 {
67 "name": "text",
68 "type": "STRING",
69 "links": [
70 11
71 ]
72 },
73 {
74 "name": "language_code",
75 "type": "STRING",
76 "links": []
77 },
78 {
79 "name": "segments_json",
80 "type": "STRING",
81 "links": [
82 12
83 ]
84 }
85 ],
86 "properties": {
87 "Node name for S&R": "FishAudioSpeechToText"
88 },
89 "widgets_values": [
90 "",
91 true
92 ],
93 "widgets_values_named": {
94 "language": "",
95 "precise_timestamps": true
96 },
97 "color": "#432",
98 "bgcolor": "#653"
99 },
100 {
101 "id": 15,
102 "type": "SaveText",
103 "pos": [
104 80,
105 260
106 ],
107 "size": [
108 310,
109 230
110 ],
111 "flags": [],
112 "order": 3,
113 "mode": 0,
114 "inputs": [
115 {
116 "name": "text",
117 "type": "STRING",
118 "link": 11
119 }
120 ],
121 "outputs": [
122 {
123 "name": "text",
124 "type": "STRING",
125 "links": null
126 }
127 ],
128 "properties": {
129 "Node name for S&R": "SaveText"
130 },
131 "widgets_values": [
132 "ComfyUI",
133 "txt"
134 ],
135 "widgets_values_named": {
136 "filename_prefix": "ComfyUI",
137 "format": "txt"
138 }
139 },
140 {
141 "id": 16,
142 "type": "SaveText",
143 "pos": [
144 80,
145 570
146 ],
147 "size": [
148 310,
149 230
150 ],
151 "flags": [],
152 "order": 4,
153 "mode": 0,
154 "inputs": [
155 {
156 "name": "text",
157 "type": "STRING",
158 "link": 12
159 }
160 ],
161 "outputs": [
162 {
163 "name": "text",
164 "type": "STRING",
165 "links": null
166 }
167 ],
168 "properties": {
169 "Node name for S&R": "SaveText"
170 },
171 "widgets_values": [
172 "ComfyUI",
173 "json"
174 ],
175 "widgets_values_named": {
176 "filename_prefix": "ComfyUI",
177 "format": "json"
178 }
179 },
180 {
181 "id": 19,
182 "type": "MarkdownNote",
183 "pos": [
184 -1000,
185 260
186 ],
187 "size": [
188 340,
189 395.328125
190 ],
191 "flags": [],
192 "order": 1,
193 "mode": 0,
194 "inputs": [],
195 "outputs": [],
196 "properties": [],
197 "widgets_values": [
198 "## Input Asset\n\n- [fish_audio_example.mp3](https://raw.githubusercontent.com/Comfy-Org/workflow_templates/refs/heads/main/input/fish_audio_example.mp3)\n\n## Guide\n\n- `segments_json` is always output, but it is empty unless `precise_timestamps` is enabled. Turn it on to get per-segment text with start/end times for subtitles.\n- `language` is optional. The model auto-detects the language regardless of what you type; a two-letter code (e.g. `en`, `zh`) only acts as a hint for mixed-language audio.\n\n## Report Issue\n\nNote: Please update ComfyUI first ([guide](https://docs.comfy.org/installation/update_comfyui)) and prepare required models. Desktop/Cloud updates follow stable releases, so some nightly-supported models may not be available yet.\n\n- Cannot run / runtime errors: [ComfyUI/issues](https://github.com/comfyanonymous/ComfyUI/issues)\n- UI / frontend issues: [ComfyUI_frontend/issues](https://github.com/Comfy-Org/ComfyUI_frontend/issues)\n- Workflow issues: [workflow_templates/issues](https://github.com/Comfy-Org/workflow_templates/issues)"
199 ],
200 "widgets_values_named": {
201 "text": "## Input Asset\n\n- [fish_audio_example.mp3](https://raw.githubusercontent.com/Comfy-Org/workflow_templates/refs/heads/main/input/fish_audio_example.mp3)\n\n## Guide\n\n- `segments_json` is always output, but it is empty unless `precise_timestamps` is enabled. Turn it on to get per-segment text with start/end times for subtitles.\n- `language` is optional. The model auto-detects the language regardless of what you type; a two-letter code (e.g. `en`, `zh`) only acts as a hint for mixed-language audio.\n\n## Report Issue\n\nNote: Please update ComfyUI first ([guide](https://docs.comfy.org/installation/update_comfyui)) and prepare required models. Desktop/Cloud updates follow stable releases, so some nightly-supported models may not be available yet.\n\n- Cannot run / runtime errors: [ComfyUI/issues](https://github.com/comfyanonymous/ComfyUI/issues)\n- UI / frontend issues: [ComfyUI_frontend/issues](https://github.com/Comfy-Org/ComfyUI_frontend/issues)\n- Workflow issues: [workflow_templates/issues](https://github.com/Comfy-Org/workflow_templates/issues)"
202 },
203 "color": "#222",
204 "bgcolor": "#000"
205 }
206 ],
207 "links": [
208 [
209 10,
210 11,
211 0,
212 13,
213 0,
214 "AUDIO"
215 ],
216 [
217 11,
218 13,
219 0,
220 15,
221 0,
222 "STRING"
223 ],
224 [
225 12,
226 13,
227 2,
228 16,
229 0,
230 "STRING"
231 ]
232 ],
233 "groups": [],
234 "config": [],
235 "extra": {
236 "ds": {
237 "scale": 0.8537713960280017,
238 "offset": [
239 1200.6783634454525,
240 173.4531048009007
241 ]
242 },
243 "frontendVersion": "1.52.5",
244 "VHS_latentpreview": false,
245 "VHS_latentpreviewrate": 0,
246 "VHS_MetadataImage": true,
247 "VHS_KeepIntermediate": true
248 },
249 "version": 0.4
250}