From d8f1d036b2daaa79c238e9edf1c6c854f3976d4c Mon Sep 17 00:00:00 2001 From: nyx-c-language Date: Tue, 18 Mar 2025 21:37:39 +0800 Subject: [PATCH] update the vocoder of synthesize and synthesize_e2e to 4 stages, and update the READEM to be consistent with the script --- examples/csmsc/tts0/README.md | 7 +++++-- examples/csmsc/tts0/run.sh | 10 ++++++++-- 2 files changed, 13 insertions(+), 4 deletions(-) diff --git a/examples/csmsc/tts0/README.md b/examples/csmsc/tts0/README.md index 8cb163659..94fa05103 100644 --- a/examples/csmsc/tts0/README.md +++ b/examples/csmsc/tts0/README.md @@ -101,7 +101,7 @@ pwg_baker_ckpt_0.4 ```bash CUDA_VISIBLE_DEVICES=${gpus} ./local/synthesize.sh --stage 0 ${conf_path} ${train_output_path} ${ckpt_name} ``` -`--stage` controls the vocoder model during synthesis, which can be `0` or `1`, use `pwgan` or `hifigan` model as vocoder. +`--stage` controls the vocoder model during synthesis, which can be `0` or `1` or `2` or `3`, use `pwgan` or `multi band melgan` or `style melgan` or `hifigan`model as vocoder. ```text usage: synthesize.py [-h] @@ -148,9 +148,12 @@ optional arguments: output dir. ``` `./local/synthesize_e2e.sh` calls `${BIN_DIR}/../synthesize_e2e.py`, which can synthesize waveform from text file. + ```bash -CUDA_VISIBLE_DEVICES=${gpus} ./local/synthesize_e2e.sh ${conf_path} ${train_output_path} ${ckpt_name} +CUDA_VISIBLE_DEVICES=${gpus} ./local/synthesize_e2e.sh --stage 0 ${conf_path} ${train_output_path} ${ckpt_name} ``` +`--stage` controls the vocoder model during synthesis, which can be `0` or `1` or `3` or `4`, use `pwgan` or `multi band melgan` or `hifigan` or `wavernn`model as vocoder. + ```text usage: synthesize_e2e.py [-h] [--am {speedyspeech_csmsc,speedyspeech_aishell3,fastspeech2_csmsc,fastspeech2_ljspeech,fastspeech2_aishell3,fastspeech2_vctk,tacotron2_csmsc,tacotron2_ljspeech}] diff --git a/examples/csmsc/tts0/run.sh b/examples/csmsc/tts0/run.sh index 3a4f5a58b..7677b0173 100755 --- a/examples/csmsc/tts0/run.sh +++ b/examples/csmsc/tts0/run.sh @@ -27,12 +27,18 @@ if [ ${stage} -le 1 ] && [ ${stop_stage} -ge 1 ]; then fi if [ ${stage} -le 2 ] && [ ${stop_stage} -ge 2 ]; then - # synthesize, vocoder is pwgan by default stage 0, stage 1 will use hifigan as vocoder + # synthesize, vocoder is pwgan by default stage 0 + # stage 1 will use multi band melgan as vocoder + # stage 2 will use style melgan as vocoder + # stage 3 will use hifigan as vocoer CUDA_VISIBLE_DEVICES=${gpus} ./local/synthesize.sh --stage 0 ${conf_path} ${train_output_path} ${ckpt_name} || exit -1 fi if [ ${stage} -le 3 ] && [ ${stop_stage} -ge 3 ]; then - # synthesize_e2e, vocoder is pwgan by default stage 0, stage 1 will use hifigan as vocoder + # synthesize_e2e, vocoder is pwgan by default stage 0 + # stage 1 will use multi band melgan as vocoder + # stage 3 will use hifigan as vocoder + # stage 4 will use wavernn as vocoder CUDA_VISIBLE_DEVICES=${gpus} ./local/synthesize_e2e.sh --stage 0 ${conf_path} ${train_output_path} ${ckpt_name} || exit -1 fi