From 1bbe59f78be9a8355bb0bee2bc8ab14ca42ac71c Mon Sep 17 00:00:00 2001 From: historia Date: Fri, 11 Sep 2026 22:03:43 -0400 Subject: Add ClearVoice MossFormer2 denoiser backend and static gap trimming --- README.md | 29 ++++++++++++++++++++++------- 1 file changed, 22 insertions(+), 7 deletions(-) (limited to 'README.md') diff --git a/README.md b/README.md index 63f5944..1d43285 100644 --- a/README.md +++ b/README.md @@ -29,7 +29,7 @@ constant bitrate at 192 kb/s by default. Loudness is normalized to **-19 LUFS with a -1.5 dBTP true-peak ceiling**. ### Examples - +``` # Optional MP3 alongside the WAV ./producer.sh recording.wav --mp3 @@ -47,18 +47,21 @@ LUFS with a -1.5 dBTP true-peak ceiling**. ./producer.sh recording.wav --denoiser fft ./producer.sh recording.wav --no-denoise +# Use the ClearVoice MossFormer2_SE_48K denoiser instead of DeepFilterNet3 +# (larger download; preserves word attacks noticeably better at high strength) +./producer.sh recording.wav --denoiser clearvoice +./producer.sh setup --denoiser clearvoice --device cpu + +# Shorten long gaps between takes (speech is never attenuated) +./producer.sh recording.wav --trim-silence --max-gap-s 1.5 --gap-pad-s 0.6 + # LibriVox-style delivery file: 44.1 kHz, 128 kbps CBR, -19 LUFS, mono ./producer.sh recording.wav \ --sample-rate 44100 --mp3 --mp3-bitrate 128 \ --target-lufs -19 --true-peak-db -1.5 ``` -All of these settings are configurable; run `./producer.sh config` to see every -setting and its default, or `--help` on any command for the flags. A -`config.toml` in the same directory as `producer.sh` is loaded automatically: -`./producer.sh config` then shows those resolved values, and an explicit -`--config FILE` or any CLI flag still overrides it. The shipped `config.toml` -documents every setting. +Edit default settings in `config.toml` ## Profiles @@ -69,6 +72,14 @@ documents every setting. | `radio` | Stronger compression, more warmth/presence, tighter range | | `cleanup-only` | Denoise + highpass only; no EQ, dynamics or normalization | +AI denoisers soften word attacks to some degree; `clearvoice` +(MossFormer2_SE_48K) was measured to leave onsets essentially untouched, +while `deepfilter` ducks them a couple of percent even at reduced strength. +Denoisers also never reach true digital silence: a noise floor remains in +blank sections. `--trim-silence` addresses both properly - long gaps are +shortened to `--gap-pad-s` of untouched room tone with a sample-exact cut, +so speech is never gated or attenuated. + Test the difference on your own audio before committing: ```bash @@ -83,6 +94,10 @@ Test the difference on your own audio before committing: driver, then retry `./producer.sh setup --device cuda`. Or use `--device cpu`. - **Sounds watery or thin:** lower the denoise strength (`--denoise-strength 0.65`) or compare profiles with `preview`. +- **Word attacks sound soft or ramp up:** AI denoisers can soften consonant + onsets. Try `--denoiser clearvoice` (larger non-causal model that sees more + attack context), lower `--denoise-strength`, or compare against + `--no-denoise` to locate the culprit. ## License -- cgit v1.2.3