test / train_config.json
saurabh1043's picture
Upload 7 files
5a1eeab verified
{
"output_dir": "/home/user_1/tts-dev/indrivoice/raya-tts/hi_en/hi/omnivoice/exp_distill_hi_8step",
"data_config": "/home/user_1/tts-dev/indrivoice/raya-tts/hi_en/hi/omnivoice/data_config_hi_dev_small.json",
"llm_name_or_path": "Qwen/Qwen3-0.6B",
"audio_vocab_size": 1025,
"audio_mask_id": 1024,
"num_audio_codebook": 8,
"audio_codebook_weights": [
8,
8,
6,
6,
4,
4,
2,
2
],
"drop_cond_ratio": 0.1,
"prompt_ratio_range": [
0.0,
0.3
],
"mask_ratio_range": [
0.0,
1.0
],
"language_ratio": 0.8,
"use_pinyin_ratio": 0.0,
"instruct_ratio": 0.0,
"only_instruct_ratio": 0.0,
"resume_from_checkpoint": null,
"init_from_checkpoint": "k2-fsa/OmniVoice",
"learning_rate": 1e-05,
"weight_decay": 0.01,
"max_grad_norm": 1.0,
"steps": 8000,
"seed": 42,
"lr_scheduler_type": "cosine",
"warmup_type": "ratio",
"warmup_ratio": 0.01,
"warmup_steps": 0,
"batch_tokens": 4096,
"gradient_accumulation_steps": 1,
"num_workers": 2,
"mixed_precision": "bf16",
"allow_tf32": true,
"use_deepspeed": false,
"deepspeed_config": null,
"attn_implementation": "sdpa",
"max_sample_tokens": 2000,
"min_sample_tokens": 50,
"max_batch_size": 32,
"logging_steps": 50,
"eval_steps": 500,
"save_steps": 500,
"keep_last_n_checkpoints": 3,
"teacher_checkpoint": "k2-fsa/OmniVoice",
"teacher_num_step": 16,
"student_num_step": 8,
"teacher_guidance_scale": 2.0,
"teacher_t_shift": 0.1,
"teacher_layer_penalty_factor": 5.0,
"teacher_position_temperature": 0.0,
"teacher_class_temperature": 0.0,
"block_size_frames": 12,
"use_block_reveal": true,
"kd_ratio": 0.5,
"kd_alpha": 0.6,
"kd_temperature": 2.0,
"kd_loss_type": "reverse_kl",
"sample_reveal_step": true,
"teacher_dtype": "float16"
}