ComfyUI Node: MW Audio Recorder

MW Audio Recorder:

The AudioRecorderSpark node is designed to facilitate audio recording and processing within the ComfyUI environment. Its primary function is to capture audio input, process it to reduce noise, and output a clean audio waveform. This node is particularly beneficial for applications requiring high-quality audio input, such as voice synthesis or audio analysis. It employs advanced techniques like spectral noise reduction and dynamic gain normalization to ensure the audio output is clear and free from unwanted noise. The node's ability to automatically detect and use silent segments as noise references makes it user-friendly and efficient, even for those with minimal technical expertise. Overall, AudioRecorderSpark is a powerful tool for anyone looking to integrate audio recording and processing into their AI-driven projects.

MW Audio Recorder Input Parameters:

trigger

The trigger parameter is a boolean that determines whether the audio recording process should start. When set to True, the node initiates the recording process; otherwise, it returns without performing any action. This parameter is crucial for controlling the execution flow and ensuring that recording only occurs when desired.

record_sec

The record_sec parameter specifies the duration of the audio recording in seconds. It directly impacts the length of the captured audio and should be set according to the user's needs. There is no explicit minimum or maximum value provided, but it should be a positive integer to ensure a valid recording duration.

n_fft

The n_fft parameter defines the number of FFT (Fast Fourier Transform) components used in the spectral analysis of the audio. It affects the resolution of the frequency analysis, with higher values providing more detailed frequency information. This parameter is essential for the noise reduction process, as it influences the accuracy of the noise profile calculation.

sensitivity

The sensitivity parameter controls the threshold for noise reduction. It determines how aggressively the node will filter out noise from the audio signal. A higher sensitivity value results in more noise being removed, but it may also affect the quality of the desired audio. Users should adjust this parameter based on the noise level in their environment.

smooth

The smooth parameter is used to define the size of the smoothing kernel applied to the noise reduction mask. It helps in preserving the edges of the audio signal while reducing noise. A larger value results in a smoother mask, which can be beneficial for maintaining audio quality during noise reduction.

sample_rate

The sample_rate parameter specifies the number of samples per second in the audio recording. It is a critical factor in determining the quality and fidelity of the recorded audio. Common sample rates include 44100 Hz and 48000 Hz, with higher rates providing better audio quality.

seed

The seed parameter is used to initialize the random number generator for any stochastic processes within the node. It ensures reproducibility of results by allowing users to generate the same output given the same input parameters and seed value.

MW Audio Recorder Output Parameters:

final_audio

The final_audio output parameter is a dictionary containing the processed audio waveform and its sample rate. The waveform is a tensor representing the cleaned audio signal, while the sample rate indicates the number of samples per second. This output is crucial for further audio processing or playback, providing a high-quality audio signal ready for use in various applications.

MW Audio Recorder Usage Tips:

• Ensure the trigger parameter is set to True to start the recording process; otherwise, the node will not perform any action.
• Adjust the sensitivity parameter based on the noise level in your environment to achieve optimal noise reduction without compromising audio quality.
• Use a higher sample_rate for applications requiring high-fidelity audio, such as music recording or detailed audio analysis.
• Experiment with different n_fft values to find the best balance between frequency resolution and computational efficiency for your specific use case.

MW Audio Recorder Common Errors and Solutions:

Recording/processing failed: <error_message>

• Explanation: This error occurs when there is an issue during the audio recording or processing phase, possibly due to incorrect parameter values or hardware issues.
• Solution: Check all input parameters for correctness, ensure your audio input device is functioning properly, and verify that the sample rate is supported by your hardware. If the problem persists, try restarting the application or your device.