Intelligent background noise removal
Separates voice from ambient noise (fans, traffic, room echoes) in real time, preserving voice naturalness. Machine learning algorithms are trained on over 10,000 hours of podcast recordings.
Audio Software Platform
Digital audio signal processing tools, designed for voice clarity and compliance with streaming platform standards.
Separates voice from ambient noise (fans, traffic, room echoes) in real time, preserving voice naturalness. Machine learning algorithms are trained on over 10,000 hours of podcast recordings.
Analyzes and adjusts sound level to -16 LUFS integrated, with presets for Spotify, Apple Podcasts, and YouTube. Detailed compliance reports and automatic corrections to avoid episode rejection.
Non-destructive editing with real-time preview, silence trimming, dynamic compression, and one-click equalization. Batch processing for multiple episodes without compromising audio quality.
1. Place the microphone 15–20 cm from your mouth, slightly to the side to avoid plosives. 2. Use a pop filter and a shock mount to reduce handling noise. 3. Record in a room with curtains or shelves to minimize echoes.
Subscribe to the newsletter and receive practical guides for setting up audio equipment, updates on LUFS standards, and tool recommendations for quick editing. No spam, just useful content for podcast creators.
LUFS (Loudness Units relative to Full Scale) is the standardized unit that measures perceived sound level. Streaming platforms like Spotify, Apple Podcasts, and YouTube apply normalization to -16 LUFS (integrated). If your episode exceeds this value, the platform automatically reduces the volume, which can lead to unwanted compression and loss of dynamics. If it is below the level, the episode will sound quieter compared to other content. Tools like ampyfy Loudness Standardizer measure and automatically correct these parameters, providing compliance reports for each platform.
The key is using a spectral separation algorithm that analyzes the frequencies of constant noise (fans, traffic, electrical hum) and selectively attenuates them, without cutting the harmonics of the voice. ampyfy Voice Clarity Engine uses machine learning trained on thousands of hours of recordings to distinguish between voice and ambient noise. The result is noise reduction of up to 18 dB, preserving the naturalness and intelligibility of the voice. For optimal results, we recommend recording at a distance of 15–20 cm from the microphone, in an acoustically treated room.
1. Position the microphone 15–20 cm from your mouth, slightly to the side (at a 45-degree angle) to avoid direct blowing and plosive consonants (p, b, t). Use a pop filter. 2. Choose a microphone with a cardioid pattern (e.g., Shure SM7B, Rode NT1) that rejects sound from the back and sides, reducing ambient noise. 3. Treat early reflections: place sound-absorbing panels (or even a thick blanket) behind the microphone and on the side walls, at a distance of 30–50 cm. This reduces echo and improves recording clarity.
An effective starting point is a three-band EQ: reduce frequencies below 80 Hz (roll-off) to eliminate low-frequency noise (traffic, hum); apply a slight boost (2–3 dB) around 3–5 kHz for clarity and presence; cut slightly (1–2 dB) around 200–300 Hz to reduce the feeling of "muddiness" or nasality. Fine adjustments depend on each speaker's voice and the recording environment. ampyfy Rapid Edit Suite includes EQ presets for different voice types and podcast genres.
Use a real-time LUFS measurement tool, integrated into your DAW or as a plugin. ampyfy Loudness Standardizer provides a detailed report including Integrated LUFS, Short-term LUFS, True Peak, and LRA (Loudness Range) values. Check that Integrated LUFS is between -17 and -15 (target -16), and True Peak does not exceed -1 dBTP (to avoid clipping during conversion). If the values are out of range, apply automatic correction or manually adjust the level with a compressor/limiter.
Dynamic compression reduces the range between the quietest and loudest parts of the recording, applying gradual attenuation above a set threshold (e.g., ratio 3:1). It is useful for leveling voice volume, especially when the speaker moves away from the microphone. Limiting, on the other hand, is compression with a very high ratio (10:1 or more) that stops any signal from exceeding a fixed threshold (e.g., -1 dBTP). It is used to prevent clipping and ensure compliance with True Peak standards. ampyfy Rapid Edit Suite includes both tools with presets specific to podcasts.
Still have questions? Contact the ampyfy team at info@ampyfy.com or call 0365019109.