ComfyUI Node: Qwen3-TTS Custom Voice

Qwen3-TTS Custom Voice:

The Qwen3CustomVoice node is designed to generate custom voice outputs using the Qwen3-TTS model. This node allows you to create personalized voice outputs by specifying various parameters such as text, language, and speaker characteristics. It is particularly beneficial for users who want to produce unique and tailored audio content, as it leverages advanced text-to-speech technology to deliver high-quality and natural-sounding voice outputs. The primary goal of this node is to provide flexibility and customization in voice generation, making it an essential tool for AI artists looking to enhance their audio projects with distinctive vocal elements.

Qwen3-TTS Custom Voice Input Parameters:

text

The text parameter is the input text that you want to convert into speech. It serves as the primary content that the node will process to generate the audio output. The quality and clarity of the generated voice depend significantly on the text provided, so it's important to ensure that the text is well-structured and free of errors. There are no specific minimum or maximum values for this parameter, but the length of the text may affect processing time and output quality.

language

The language parameter specifies the language in which the text should be spoken. This parameter is crucial for ensuring that the generated voice accurately reflects the linguistic characteristics of the desired language. The available options for this parameter depend on the languages supported by the Qwen3-TTS model. Selecting the correct language is essential for achieving a natural and intelligible voice output.

target_speaker

The target_speaker parameter allows you to choose a specific speaker profile for the voice generation. This parameter is important for customizing the voice characteristics, such as tone and pitch, to match your desired speaker identity. The options for this parameter are typically predefined speaker profiles available within the model. Choosing the right speaker profile can significantly enhance the personalization of the generated voice.

instruct

The instruct parameter provides additional instructions or context for the voice generation process. This parameter can be used to influence the style or emotion of the generated voice, allowing for more expressive and contextually appropriate audio outputs. The specific instructions you can provide depend on the capabilities of the Qwen3-TTS model.

max_new_tokens

The max_new_tokens parameter defines the maximum number of tokens that the model can generate for the voice output. This parameter helps control the length of the generated audio, ensuring that it does not exceed a certain duration. Adjusting this parameter can be useful for managing processing time and ensuring that the output fits within specific time constraints.

Qwen3-TTS Custom Voice Output Parameters:

audio

The audio parameter is the primary output of the Qwen3CustomVoice node, representing the generated voice in audio format. This output is the result of processing the input text and parameters through the Qwen3-TTS model, producing a high-quality and natural-sounding voice that reflects the specified characteristics. The audio output can be used in various applications, such as voiceovers, podcasts, or any project requiring custom voice content.

Qwen3-TTS Custom Voice Usage Tips:

• Ensure that the input text is clear and well-structured to achieve the best audio quality.
• Experiment with different speaker profiles and instructions to find the most suitable voice characteristics for your project.
• Adjust the max_new_tokens parameter to control the length of the generated audio, especially for longer texts.

Qwen3-TTS Custom Voice Common Errors and Solutions:

Model Type Error: You are trying to use 'Custom Voice' with an incompatible model.

• Explanation: This error occurs when you attempt to use the Qwen3CustomVoice node with a model that does not support custom voice generation.
• Solution: Load a compatible model, such as Qwen3-TTS-12Hz-1.7B-CustomVoice, to ensure that the node functions correctly.

ValueError: Text input is empty or invalid.

• Explanation: This error indicates that the input text provided to the node is either empty or not properly formatted.
• Solution: Verify that the input text is non-empty and correctly formatted before processing it through the node.