bambara-tts-2 / config.json
oza75's picture
Upload Xtts
d47b240 verified
raw
history blame
No virus
1.32 kB
{
"_name_or_path": "./bm_xtts",
"architectures": [
"Xtts"
],
"clvp_checkpoint": null,
"cond_d_vector_in_each_upsampling_layer": true,
"d_vector_dim": 512,
"debug_loading_failures": false,
"decoder_checkpoint": null,
"decoder_input_dim": 1024,
"duration_const": 102400,
"dvae_checkpoint": "",
"enable_redaction": false,
"gpt_batch_size": 1,
"gpt_checkpoint": null,
"gpt_code_stride_len": 1024,
"gpt_layers": 30,
"gpt_loss_mel_ce_weight": 1.0,
"gpt_loss_text_ce_weight": 0.01,
"gpt_max_audio_tokens": 605,
"gpt_max_prompt_tokens": 70,
"gpt_max_text_tokens": 402,
"gpt_n_heads": 16,
"gpt_n_model_channels": 1024,
"gpt_num_audio_tokens": 1026,
"gpt_number_text_tokens": 50257,
"gpt_start_audio_token": 1024,
"gpt_start_text_token": 50259,
"gpt_stop_audio_token": 1025,
"gpt_stop_text_token": 50260,
"gpt_use_masking_gt_prompt_approach": true,
"gpt_use_perceiver_resampler": true,
"input_sample_rate": 22050,
"kv_cache": true,
"max_text_length": 200,
"model_type": "xtts",
"num_chars": 255,
"output_hop_length": 256,
"output_sample_rate": 24000,
"torch_dtype": "float32",
"transformers_version": "4.41.2",
"vocoder": "",
"vq_loss_weight": 1.2,
"vqvae_checkpoint": "oza75/bambara-vqvae",
"wav_loss_weight": 1.5,
"xtts_checkpoint": ""
}