tasks/gen_batch_e2e_server.log

Returning existing local_dir `/home/barberb/IndexTTS-2-Demo/checkpoints` as remote repo cannot be accessed in `snapshot_download` (Cannot reach https://huggingface.co/api/models/IndexTeam/IndexTTS-2/revision/main: offline mode is enabled. To disable it, please unset the `HF_HUB_OFFLINE` environment variable.).
GPT2InferenceModel has generative capabilities, as `prepare_inputs_for_generation` is explicitly overwritten. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
/home/barberb/IndexTTS-2-Demo/indextts/utils/front.py:113: RuntimeWarning: Text normalization dependencies unavailable; using passthrough normalization. Original error: No module named 'tn'
  warnings.warn(
Passing a tuple of `past_key_values` is deprecated and will be removed in Transformers v4.53.0. You should pass an instance of `Cache` instead, e.g. `past_key_values=DynamicCache.from_legacy_cache(past_key_values)`.
[HuggingFace] Downloading models to /home/barberb/IndexTTS-2-Demo/checkpoints,model cache dir=/home/barberb/IndexTTS-2-Demo/checkpoints/hf_cache
>> Be patient, it may take a while to run in CPU mode.
>> GPT weights restored from: /home/barberb/IndexTTS-2-Demo/checkpoints/gpt.pth
>> semantic_codec weights restored from: ./checkpoints/hf_cache/models--amphion--MaskGCT/snapshots/265c6cef07625665d0c28d2faafb1415562379dc/semantic_codec/model.safetensors
cfm loaded
length_regulator loaded
gpt_layer loaded
>> s2mel weights restored from: /home/barberb/IndexTTS-2-Demo/checkpoints/s2mel.pth
>> campplus_model weights restored from: ./checkpoints/hf_cache/models--funasr--campplus/snapshots/81a8afba4ca420cf6f845f157d5fc1d365286821/campplus_cn_common.bin
Loading config.json from local directory
Loading weights from local directory
Removing weight norm...
>> bigvgan weights restored from: ./checkpoints/hf_cache/models--nvidia--bigvgan_v2_22khz_80band_256x/snapshots/633ff708ed5b74903e86ff1298cf4a98e921c513
>> TextNormalizer loaded
>> bpe model loaded from: /home/barberb/IndexTTS-2-Demo/checkpoints/bpe.model
* Running on local URL:  http://127.0.0.1:7861
* To create a public link, set `share=True` in `launch()`.
Emo control mode:0,weight:0.8,vec:None
>> starting inference...
Use the specified emotion vector

  0%|          | 0/25 [00:00<?, ?it/s]
  4%|▍         | 1/25 [00:00<00:16,  1.43it/s]
  8%|▊         | 2/25 [00:01<00:15,  1.46it/s]
 12%|█▏        | 3/25 [00:02<00:15,  1.46it/s]
 16%|█▌        | 4/25 [00:02<00:14,  1.47it/s]
 20%|██        | 5/25 [00:03<00:13,  1.49it/s]
 24%|██▍       | 6/25 [00:04<00:12,  1.47it/s]
 28%|██▊       | 7/25 [00:04<00:12,  1.46it/s]
 32%|███▏      | 8/25 [00:05<00:11,  1.47it/s]
 36%|███▌      | 9/25 [00:06<00:10,  1.48it/s]
 40%|████      | 10/25 [00:06<00:10,  1.49it/s]
 44%|████▍     | 11/25 [00:07<00:09,  1.47it/s]
 48%|████▊     | 12/25 [00:08<00:08,  1.48it/s]
 52%|█████▏    | 13/25 [00:08<00:08,  1.49it/s]
 56%|█████▌    | 14/25 [00:09<00:07,  1.48it/s]
 60%|██████    | 15/25 [00:10<00:06,  1.47it/s]
 64%|██████▍   | 16/25 [00:10<00:06,  1.47it/s]
 68%|██████▊   | 17/25 [00:11<00:05,  1.48it/s]
 72%|███████▏  | 18/25 [00:12<00:04,  1.48it/s]
 76%|███████▌  | 19/25 [00:12<00:04,  1.47it/s]
 80%|████████  | 20/25 [00:13<00:03,  1.47it/s]
 84%|████████▍ | 21/25 [00:14<00:02,  1.46it/s]
 88%|████████▊ | 22/25 [00:14<00:02,  1.46it/s]
 92%|█████████▏| 23/25 [00:15<00:01,  1.47it/s]
 96%|█████████▌| 24/25 [00:16<00:00,  1.46it/s]
100%|██████████| 25/25 [00:16<00:00,  1.46it/s]
100%|██████████| 25/25 [00:16<00:00,  1.47it/s]
>> gpt_gen_time: 83.86 seconds
>> gpt_forward_time: 3.20 seconds
>> s2mel_time: 17.02 seconds
>> bigvgan_time: 2.91 seconds
>> Total inference time: 110.27 seconds
>> Generated audio length: 2.79 seconds
>> [batch] batch_num: 1 bucket_max_size: 1 bucket_count: 1
>> RTF: 39.5747
>> wav file saved to: outputs/spk_1779909641-item-1.wav
>> starting inference...
Use the specified emotion vector

  0%|          | 0/25 [00:00<?, ?it/s]
  4%|▍         | 1/25 [00:00<00:20,  1.17it/s]
  8%|▊         | 2/25 [00:01<00:19,  1.16it/s]
 12%|█▏        | 3/25 [00:02<00:19,  1.15it/s]
 16%|█▌        | 4/25 [00:03<00:18,  1.16it/s]
 20%|██        | 5/25 [00:04<00:17,  1.15it/s]
 24%|██▍       | 6/25 [00:05<00:16,  1.13it/s]
 28%|██▊       | 7/25 [00:06<00:15,  1.13it/s]
 32%|███▏      | 8/25 [00:07<00:15,  1.13it/s]
 36%|███▌      | 9/25 [00:07<00:14,  1.13it/s]
 40%|████      | 10/25 [00:08<00:13,  1.13it/s]
 44%|████▍     | 11/25 [00:09<00:12,  1.12it/s]
 48%|████▊     | 12/25 [00:10<00:11,  1.12it/s]
 52%|█████▏    | 13/25 [00:11<00:10,  1.12it/s]
 56%|█████▌    | 14/25 [00:12<00:09,  1.12it/s]
 60%|██████    | 15/25 [00:13<00:08,  1.11it/s]
 64%|██████▍   | 16/25 [00:14<00:08,  1.12it/s]
 68%|██████▊   | 17/25 [00:15<00:07,  1.13it/s]
 72%|███████▏  | 18/25 [00:15<00:06,  1.14it/s]
 76%|███████▌  | 19/25 [00:16<00:05,  1.14it/s]
 80%|████████  | 20/25 [00:17<00:04,  1.14it/s]
 84%|████████▍ | 21/25 [00:18<00:03,  1.13it/s]
 88%|████████▊ | 22/25 [00:19<00:02,  1.13it/s]
 92%|█████████▏| 23/25 [00:20<00:01,  1.11it/s]
 96%|█████████▌| 24/25 [00:21<00:00,  1.12it/s]
100%|██████████| 25/25 [00:22<00:00,  1.11it/s]
100%|██████████| 25/25 [00:22<00:00,  1.13it/s]