open-mmlab / Amphion

Amphion (/æmˈfaɪən/) is a toolkit for Audio, Music, and Speech Generation. Its purpose is to support reproducible research and help junior researchers and engineers get started in the field of audio, music, and speech generation research and development.
https://openhlt.github.io/amphion/
MIT License
7.81k stars 590 forks source link

[Help]: TTA 推理缺失的配置文件 #332

Closed cpken closed 3 weeks ago

cpken commented 3 weeks ago

请提供 TTA 推理所需的文件的下载链接。

egs/tta/audioldm/run_inference.sh

######## Set Experiment Configuration ###########
exp_config="$exp_dir/exp_config.json"
exp_name="audioldm_debug_latent_size_4_5_39"
checkpoint_path="$work_dir/ckpts/tta/audioldm_debug_latent_size_4_5_39/checkpoints/step-0570000_loss-0.2521.pt"
output_dir="$work_dir/temp"
vocoder_config_path="$work_dir/ckpts/tta/hifigan_checkpoints/config.json"
vocoder_path="$work_dir/ckpts/tta/hifigan_checkpoints/g_01250000"
num_steps=200
guidance_scale=4.0
cpken commented 3 weeks ago

模型下载地址 https://www.modelscope.cn/models/amphion/text_to_audio/summary