Hojo-TTS-Light / models /README.md
inoryQwQ's picture
Add 20s long demo audio (en/zh, sampling 0.8/0.95/rep1.1); post_config sampling default for long-text stability
b2e96cd verified
|
Raw
History Blame Contribute Delete
1.02 kB

模型资产

组件 文件 说明
LM (10 层 decode-only) lm_s8/qwen3_p8_l0..9_together.axmodel + qwen3_post.axmodel + embed_tokens.bin + config.json + post_config.json + tokenizer.* ax-llm s8,Pulsar2 7.0 llm_build2,kv_cache_len=2176;post_config 采样 0.8/0.95/rep 1.1(长文本防重复)
fine_local fine_local.axmodel INT8,cos 0.9997
decoder decoder_sq.axmodel SmoothQuant+U16,mag 0.998+/phase 0.95-0.97
词典 tokenizer.json / tokenizer_config.json 原版 Hojo
音色 Hojo-TTS-Light-40M-voice.npz token_embedding + speaker_embeds/vecs(15 音色)

prompt embedding 生成(host 端)

python tools/prepare_prompt_embeds.py \
  --text "Hello, this is a demo from Hojo TTS Light." --voice 9 \
  --models-dir origin/models --out-prefix demo_en
# 产物: demo_en.bin(BF16 [num_tokens,512])+ demo_en.meta.json

板端部署路径示例:LM → /mnt/sdcard/hojo_axllm/model_s8,其余 → /mnt/sdcard/hojo_e2e。