playmore/speech-enhancer 🖼️❓✓ → 🖼️

▶️ 603 runs 📅 Nov 2025 ⚙️ Cog 0.16.9
audio-denoising audio-to-audio speech-enhancement

About

Speech Enhancer

Example Output

Output

Example output

Performance Metrics

2.08s Prediction Time
2.10s Total Time
All Input Parameters 
{
  "audio": "https://replicate.delivery/pbxt/O5zFqg3CK7UIAvnKo3atcq5ahAK2iQVSxEGHaZ7lWyR27MSZ/cafe.wav",
  "model": "mossformer2_se_48k",
  "batch_process_chunks": false,
  "force_sliding_window": false
}
Input Parameters
audio (required) Type: string
Noisy audio file (wav/mp3/m4a/etc.) to denoise.
model Default: mossformer2_se_48k
Choose a model to run.
batch_process_chunks Type: booleanDefault: false
Only applies to resemble_denoiser; process 30-second chunks in batches for throughput.
force_sliding_window Type: booleanDefault: false
Only applies to mossformer2; otherwise ignored. Forces sliding-window decoding regardless of clip length.
Output Schema

Output

Type: string • Format: uri

Example Execution Logs 
/src/denoise_app/mossformer2/features.py:70: UserWarning: The length of signal is shorter than the length parameter. Result is being padded with zeros in the tail. Please check your center and hop_length settings. (Triggered internally at ../aten/src/ATen/native/SpectralOps.cpp:1187.)
return torch.istft(
Version Details
Version ID
bda37cf8cb38f5b677514933634a281b263a04225f7b2bf62c1c1b8748d21ae6
Version Created
November 20, 2025
Run on Replicate →