토킹헤드 비디오 생성
Talkvid ID-LoRA와 함께 Qwen 3 TTS 및 LTX 2.3으로 오픈 소스 비디오 생성 파이프라인에서 목소리 일관성을 유지하세요.
워크플로 다운로드Cloud에서 사용해 보기 API 문서
00:00
워크플로 내부
소스 템플릿의 읽기 전용 미리보기입니다. Cloud에서 열어 노드를 살펴보고 변경할 수 있습니다.
코드에서 이 모델 호출하기
아래 요청은 Playground와 동일하게 준비된 그래프와 설정을 사용합니다. 제출할 때마다 유료 실행이 시작될 수 있으므로 결과를 알 수 없는 제출을 자동으로 재시도하지 마세요.
1 API 키 가져오기 ↗
What you need
- Runs on
- Comfy Cloud
- 키
- COMFY_API_KEY
- 내 파일
- 코드가 해당 URL에서 업로드
유료 Cloud 요금제와 크레딧이 필요합니다.
API JSON 다운로드API 문서 ↗
# Python 3.10+: pip install comfy-sdk==0.4.0
import os
from comfy_sdk import Comfy
api_key = os.environ.get("COMFY_API_KEY", "").strip()
if not api_key: raise RuntimeError("Set COMFY_API_KEY")
os.environ.setdefault("COMFY_BASE_URL", "https://cloud.comfy.org")
client = Comfy(api_key=api_key)
workflow = client.workflows.from_json({
"440": {
"class_type": "LoadImage",
"inputs": {
"image": "UPLOADED_image_FILENAME"
},
"_meta": {
"title": "Load Image"
}
},
"5432": {
"class_type": "SaveVideo",
"inputs": {
"filename_prefix": "video/ComfyUI",
"format": "auto",
"format.codec": "auto",
"video": [
"5436:653",
0
]
},
"_meta": {
"title": "Save Video"
}
},
"5434": {
"class_type": "FB_Qwen3TTSCustomVoice",
"inputs": {
"attention": "auto",
"custom_model_path": "",
"custom_speaker_name": "",
"device": "cuda",
"instruct": [
"5439",
0
],
"language": "Auto",
"max_new_tokens": 2048,
"model_choice": "1.7B",
"precision": "bf16",
"repetition_penalty": 1.05,
"seed": 279848348572770,
"speaker": "Dylan",
"temperature": 1,
"text": [
"5440",
0
],
"top_k": 20,
"top_p": 0.8,
"unload_model_after_generate": False
},
"_meta": {
"title": "🎵 Qwen3-TTS CustomVoice"
}
},
"5435": {
"class_type": "SaveAudioMP3",
"inputs": {
"audio": [
"5434",
0
],
"filename_prefix": "audio/ComfyUI",
"quality": "V0"
},
"_meta": {
"title": "Save Audio (MP3) (DEPRECATED)"
}
},
"5437": {
"class_type": "PreviewAny",
"inputs": {
"source": [
"5441",
0
]
},
"_meta": {
"title": "Preview as Text"
}
},
"5438": {
"class_type": "RegexExtract",
"inputs": {
"case_insensitive": True,
"dotall": False,
"group_index": 1,
"mode": "First Group",
"multiline": False,
"regex_pattern": "(?s)(?<=\\n---\\n)([\\s\\S]*)$",
"string": [
"5441",
0
]
},
"_meta": {
"title": "Extract Text"
}
},
"5439": {
"class_type": "PreviewAny",
"inputs": {
"source": [
"5438",
0
]
},
"_meta": {
"title": "Preview as Text"
}
},
"5440": {
"class_type": "PreviewAny",
"inputs": {
"source": [
"5442",
0
]
},
"_meta": {
"title": "Preview as Text"
}
},
"5441": {
"class_type": "GeminiNode",
"inputs": {
"images": [
"440",
0
],
"model": "gemini-3-1-pro",
"prompt": "",
"seed": 273799329880563,
"system_prompt": "Role: You are a professional prompt writer for ID-LoRA prompt generation.\n\nTask: Generate a two-part output from the user’s input.\n\nStrict Output Constraint (IMPORTANT):\n\nReturn the output as RAW PLAIN TEXT only.\nDO NOT use Markdown code blocks.\nDO NOT use any headers, labels, or bold text outside of the generated two-part output.\nUse ONLY the separator --- to divide the two parts.\nPart 1: spoken dialogue/performance text only, optimized to sound natural when spoken aloud.\nPart 2: the same concept rewritten in the exact ID-LoRA tagged format using these three sections only:\n[VISUAL]: ...\n[SPEECH]: ...\n[SOUNDS]: ...\nDo NOT output anything before Part 1 or after Part 2.\n\nGlobal Length Rule:\n\nKeep the spoken content short enough to produce about 10 seconds of speech unless the user explicitly asks otherwise.\nAim for roughly 20 to 35 spoken words in Part 1.\nCondense long inputs aggressively while preserving the core meaning, tone, and key message.\nPrioritize brevity, clarity, speakability, and a strong opening hook over completeness.\nRemove repetition, side details, filler, and nonessential context.\nIf the input is too long, compress it into a concise spoken version rather than preserving everything.\n\nPart 1 Rules:\n\nPart 1 must contain ONLY the words meant to be spoken aloud.\nDo NOT begin Part 1 with visual description, scene-setting narration, character description, or action description.\nStart immediately with the most attention-grabbing, high-impact spoken hook.\nThe hook should appear as early as possible, ideally in the first sentence or phrase.\nDo NOT describe what the character looks like, where they are, what they are wearing, what they are holding, or what is happening visually unless that information is spoken by the character as part of the dialogue itself.\nPart 1 should read like a real spoken performance, monologue, ad read, or direct address.\nUse punctuation, pauses, ellipses, and selective capitalization to improve spoken delivery when useful.\nDo NOT use SSML.\nDo NOT use non-voice tags such as music, ambience, sound effects, camera notes, or technical markup.\nPrefer one clear hook, one core message or benefit, and one short closing beat.\n\nPart 2 Rules:\n\nPart 2 must contain exactly three tagged sections in this exact order:\n[VISUAL]:\n[SPEECH]:\n[SOUNDS]:\n\n[VISUAL] Rules:\nDescribe the shot type, subject appearance, clothing, setting, lighting, framing, and visible actions.\nBe descriptive enough to guide generation clearly, but keep it compact and production-useful.\nExplicitly indicate that the person is speaking or talking to camera so the line is generated as on-screen speech rather than voice-over.\nDefault toward believable UGC-style behavior unless the user asks otherwise:\ndirect-to-camera delivery\nhandheld or phone-like framing\nselfie or testimonial feel\nnatural gestures\nauthentic facial expression\nslight body movement\ncasual presenting or showing when relevant\nAvoid cinematic, polished, theatrical, or overly staged action unless the user explicitly asks for that.\n\n[SPEECH] Rules:\nWrite the exact words the person should say.\nThis must be the literal transcript, not a summary.\nKeep it closely aligned with Part 1, ideally verbatim except for minor punctuation cleanup if needed.\nDo NOT add scene description, camera notes, or sound cues inside [SPEECH].\n\n[SOUNDS] Rules:\nDescribe both the vocal delivery and the ambient/background audio.\nInclude speaker qualities such as tone, volume, pace, energy, and mic proximity or distance.\nInclude relevant environmental sounds, room tone, music, nature sounds, or other ambience when appropriate.\nKeep the audio grounded, coherent with the visual scene, and not overly busy unless requested.\n\nConsistency Rules:\n\nPart 1 and Part 2 must describe the same idea, message, tone, and scene.\n[SPEECH] in Part 2 should match Part 1 as closely as possible.\n[VISUAL], [SPEECH], and [SOUNDS] must feel like one unified prompt, not separate concepts.\nDo not introduce unrelated ideas, props, settings, or actions that were not implied by the user’s request.\nIf the user provides exact wording, preserve it in Part 1 and [SPEECH] unless the user asks for rewriting or shortening.\n\nSafety and Quality Rules:\n\nDo not introduce sensitive, explicit, hateful, political, or unsafe content that was not already in the user input.\nEnsure the final output always contains exactly two parts separated by ---.\nEnsure Part 2 always uses the exact three ID-LoRA tags and includes all three of them once."
},
"_meta": {
"title": "Google Gemini"
}
},
"5442": {
"class_type": "RegexExtract",
"inputs": {
"case_insensitive": True,
"dotall": False,
"group_index": 1,
"mode": "First Group",
"multiline": False,
"regex_pattern": "(?sm)^(.*?)^---$(.*)",
"string": [
"5441",
0
]
},
"_meta": {
"title": "Extract Text"
}
},
"5436:293": {
"class_type": "LoraLoaderModelOnly",
"inputs": {
"lora_name": "ltx-2.3-22b-distilled-lora-384.safetensors",
"model": [
"5436:317",
0
],
"strength_model": 0.5
},
"_meta": {
"title": "Load LoRA"
}
},
"5436:313": {
"class_type": "LatentUpscaleModelLoader",
"inputs": {
"model_name": "ltx-2.3-spatial-upscaler-x2-1.1.safetensors"
},
"_meta": {
"title": "Load Latent Upscale Model"
}
},
"5436:317": {
"class_type": "CheckpointLoaderSimple",
"inputs": {
"ckpt_name": "ltx-2.3-22b-dev-fp8.safetensors"
},
"_meta": {
"title": "Load Checkpoint"
}
},
"5436:318": {
"class_type": "LTXAVTextEncoderLoader",
"inputs": {
"ckpt_name": "ltx-2.3-22b-dev-fp8.safetensors",
"device": "default",
"text_encoder": "gemma_3_12B_it_fp4_mixed.safetensors"
},
"_meta": {
"title": "Load LTXV Audio Text Encoder"
}
},
"5436:335": {
"class_type": "LTXVAudioVAELoader",
"inputs": {
"ckpt_name": "ltx-2.3-22b-dev-fp8.safetensors"
},
"_meta": {
"title": "Load LTXV Audio VAE"
}
},
"5436:346": {
"class_type": "LoraLoaderModelOnly",
"inputs": {
"lora_name": "ltx-2.3-id-lora-celebvhq-3k.safetensors",
"model": [
"5436:293",
0
],
"strength_model": 1
},
"_meta": {
"title": "Load LoRA"
}
},
"5436:348": {
"class_type": "LTXVEmptyLatentAudio",
"inputs": {
"audio_vae": [
"5436:335",
0
],
"batch_size": 1,
"frame_rate": [
"5436:652",
1
],
"frames_number": [
"5436:657",
1
]
},
"_meta": {
"title": "Empty LTXV Latent Audio"
}
},
"5436:349": {
"class_type": "LTXVReferenceAudio",
"inputs": {
"audio_vae": [
"5436:335",
0
],
"end_percent": 1,
"identity_guidance_scale": 3,
"model": [
"5436:346",
0
],
"negative": [
"5436:642",
0
],
"positive": [
"5436:636",
0
],
"reference_audio": [
"5434",
0
],
"start_percent": 0
},
"_meta": {
"title": "LTXV Reference Audio (ID-LoRA)"
}
},
"5436:624": {
"class_type": "RandomNoise",
"inputs": {
"noise_seed": 42
},
"_meta": {
"title": "RandomNoise"
}
},
"5436:625": {
"class_type": "RandomNoise",
"inputs": {
"noise_seed": 529262879849640
},
"_meta": {
"title": "RandomNoise"
}
},
"5436:626": {
"class_type": "LTXVConcatAVLatent",
"inputs": {
"audio_latent": [
"5436:639",
1
],
"video_latent": [
"5436:633",
0
]
},
"_meta": {
"title": "Concat AV Latent"
}
},
"5436:627": {
"class_type": "KSamplerSelect",
"inputs": {
"sampler_name": "euler_cfg_pp"
},
"_meta": {
"title": "KSamplerSelect"
}
},
"5436:628": {
"class_type": "ManualSigmas",
"inputs": {
"sigmas": "0.85, 0.7250, 0.4219, 0.0"
},
"_meta": {
"title": "ManualSigmas"
}
},
"5436:629": {
"class_type": "CFGGuider",
"inputs": {
"cfg": 1,
"model": [
"5436:293",
0
],
"negative": [
"5436:631",
1
],
"positive": [
"5436:631",
0
]
},
"_meta": {
"title": "CFG Guider"
}
},
"5436:630": {
"class_type": "SamplerCustomAdvanced",
"inputs": {
"guider": [
"5436:643",
0
],
"latent_image": [
"5436:659",
0
],
"noise": [
"5436:625",
0
],
"sampler": [
"5436:634",
0
],
"sigmas": [
"5436:638",
0
]
},
"_meta": {
"title": "SamplerCustomAdvanced"
}
},
"5436:631": {
"class_type": "LTXVCropGuides",
"inputs": {
"latent": [
"5436:639",
0
],
"negative": [
"5436:637",
1
],
"positive": [
"5436:637",
0
]
},
"_meta": {
"title": "LTXVCropGuides"
}
},
"5436:632": {
"class_type": "LTXVLatentUpsampler",
"inputs": {
"samples": [
"5436:639",
0
],
"upscale_model": [
"5436:313",
0
],
"vae": [
"5436:317",
2
]
},
"_meta": {
"title": "LTXVLatentUpsampler"
}
},
"5436:633": {
"class_type": "LTXVImgToVideoInplace",
"inputs": {
"bypass": False,
"image": [
"5436:649",
0
],
"latent": [
"5436:632",
0
],
"strength": 1,
"vae": [
"5436:317",
2
]
},
"_meta": {
"title": "LTXVImgToVideoInplace"
}
},
"5436:634": {
"class_type": "KSamplerSelect",
"inputs": {
"sampler_name": "euler_ancestral_cfg_pp"
},
"_meta": {
"title": "KSamplerSelect"
}
},
"5436:635": {
"class_type": "LTXVAudioVAEDecode",
"inputs": {
"audio_vae": [
"5436:335",
0
],
"samples": [
"5436:641",
1
]
},
"_meta": {
"title": "LTXV Audio VAE Decode"
}
},
"5436:636": {
"class_type": "CLIPTextEncode",
"inputs": {
"clip": [
"5436:318",
0
],
"text": [
"5436:660",
0
]
},
"_meta": {
"title": "CLIP Text Encode (Prompt)"
}
},
"5436:637": {
"class_type": "LTXVConditioning",
"inputs": {
"frame_rate": [
"5436:652",
0
],
"negative": [
"5436:349",
2
],
"positive": [
"5436:349",
1
]
},
"_meta": {
"title": "LTXVConditioning"
}
},
"5436:638": {
"class_type": "ManualSigmas",
"inputs": {
"sigmas": "1.0, 0.99375, 0.9875, 0.98125, 0.975, 0.909375, 0.725, 0.421875, 0.0"
},
"_meta": {
"title": "ManualSigmas"
}
},
"5436:639": {
"class_type": "LTXVSeparateAVLatent",
"inputs": {
"av_latent": [
"5436:630",
0
]
},
"_meta": {
"title": "Separate AV Latent"
}
},
"5436:640": {
"class_type": "SamplerCustomAdvanced",
"inputs": {
"guider": [
"5436:629",
0
],
"latent_image": [
"5436:626",
0
],
"noise": [
"5436:624",
0
],
"sampler": [
"5436:627",
0
],
"sigmas": [
"5436:628",
0
]
},
"_meta": {
"title": "SamplerCustomAdvanced"
}
},
"5436:641": {
"class_type": "LTXVSeparateAVLatent",
"inputs": {
"av_latent": [
"5436:640",
0
]
},
"_meta": {
"title": "Separate AV Latent"
}
},
"5436:642": {
"class_type": "CLIPTextEncode",
"inputs": {
"clip": [
"5436:318",
0
],
"text": "pc game, console game, video game, cartoon, childish, ugly"
},
"_meta": {
"title": "CLIP Text Encode (Prompt)"
}
},
"5436:643": {
"class_type": "CFGGuider",
"inputs": {
"cfg": 1,
"model": [
"5436:349",
0
],
"negative": [
"5436:637",
1
],
"positive": [
"5436:637",
0
]
},
"_meta": {
"title": "CFG Guider"
}
},
"5436:644": {
"class_type": "VAEDecodeTiled",
"inputs": {
"overlap": 64,
"samples": [
"5436:641",
0
],
"temporal_overlap": 4,
"temporal_size": 4096,
"tile_size": 768,
"vae": [
"5436:317",
2
]
},
"_meta": {
"title": "VAE Decode (Tiled)"
}
},
"5436:645": {
"class_type": "LTXVImgToVideoInplace",
"inputs": {
"bypass": False,
"image": [
"5436:649",
0
],
"latent": [
"5436:656",
0
],
"strength": 0.7,
"vae": [
"5436:317",
2
]
},
"_meta": {
"title": "LTXVImgToVideoInplace"
}
},
"5436:646": {
"class_type": "PrimitiveInt",
"inputs": {
"value": 1280
},
"_meta": {
"title": "Width"
}
},
"5436:647": {
"class_type": "SolidMask",
"inputs": {
"height": [
"5436:658",
0
],
"value": 0,
"width": [
"5436:646",
0
]
},
"_meta": {
"title": "Create Solid Mask"
}
},
"5436:648": {
"class_type": "ResizeImagesByLongerEdge",
"inputs": {
"images": [
"5436:661",
0
],
"longer_edge": 1536
},
"_meta": {
"title": "Resize Images by Longer Edge (DEPRECATED)"
}
},
"5436:649": {
"class_type": "LTXVPreprocess",
"inputs": {
"image": [
"5436:648",
0
],
"img_compression": 18
},
"_meta": {
"title": "LTXV Preprocess"
}
},
"5436:650": {
"class_type": "ComfyMathExpression",
"inputs": {
"expression": "a/2",
"values.a": [
"5436:658",
0
]
},
"_meta": {
"title": "Math Expression"
}
},
"5436:651": {
"class_type": "ComfyMathExpression",
"inputs": {
"expression": "a/2",
"values.a": [
"5436:646",
0
]
},
"_meta": {
"title": "Math Expression"
}
},
"5436:652": {
"class_type": "ComfyMathExpression",
"inputs": {
"expression": "a",
"values.a": [
"5436:655",
0
]
},
"_meta": {
"title": "Math Expression"
}
},
"5436:653": {
"class_type": "CreateVideo",
"inputs": {
"audio": [
"5436:635",
0
],
"bit_depth": "auto",
"codec": "none",
"color_space": "sRGB",
"fps": [
"5436:652",
0
],
"images": [
"5436:644",
0
]
},
"_meta": {
"title": "Create Video"
}
},
"5436:654": {
"class_type": "PrimitiveFloat",
"inputs": {
"value": 9
},
"_meta": {
"title": "Duration"
}
},
"5436:655": {
"class_type": "PrimitiveInt",
"inputs": {
"value": 25
},
"_meta": {
"title": "Frame Rate"
}
},
"5436:656": {
"class_type": "EmptyLTXVLatentVideo",
"inputs": {
"batch_size": 1,
"height": [
"5436:650",
1
],
"length": [
"5436:657",
1
],
"width": [
"5436:651",
1
]
},
"_meta": {
"title": "Empty LTXV Latent Video"
}
},
"5436:657": {
"class_type": "ComfyMathExpression",
"inputs": {
"expression": "a * b + 1",
"values.a": [
"5436:654",
0
],
"values.b": [
"5436:655",
0
]
},
"_meta": {
"title": "Math Expression"
}
},
"5436:658": {
"class_type": "PrimitiveInt",
"inputs": {
"value": 720
},
"_meta": {
"title": "Height"
}
},
"5436:659": {
"class_type": "LTXVConcatAVLatent",
"inputs": {
"audio_latent": [
"5436:348",
0
],
"video_latent": [
"5436:645",
0
]
},
"_meta": {
"title": "Concat AV Latent"
}
},
"5436:660": {
"class_type": "PrimitiveStringMultiline",
"inputs": {
"value": [
"5439",
0
]
},
"_meta": {
"title": "Prompt"
}
},
"5436:661": {
"class_type": "ResizeImageMaskNode",
"inputs": {
"input": [
"440",
0
],
"resize_type": "scale dimensions",
"resize_type.crop": "center",
"resize_type.height": [
"5436:658",
0
],
"resize_type.width": [
"5436:646",
0
],
"scale_method": "lanczos"
},
"_meta": {
"title": "Resize Image/Mask"
}
}
})
workflow.set_input("440", "image", client.assets.from_url("https://comfy.org/catalog-media/f7a45733-82ac-4831-8d39-a436efc22421/91e50b9e-fa45-46e0-989d-9e798bc94aa9"))
# run() also passes api_key to the partner nodes in this workflow.
job = client.run(workflow, api_key=api_key)
for output in job.get_outputs("5432"):
output.to_file(output.name)
for output in job.get_outputs("5435"):
output.to_file(output.name)