Downloads · 30 days
0
melo-tts-rknn
voicialex/melo-tts-rknn
melo-tts-rknn is a machine learning model from voicialex. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
Chinese/English TTS (Text-to-Speech) — MeloTTS converted for RK3588 NPU.
Access
Public
Updated Jun 8, 2026
Repo size
526 MB
Likes
1
Public
Repo makeup
Click a slice to open those files.
.rknn526 MB · 98%
At a glance
- Access
- Public
- Created
- Jun 8, 2026
- Updated
- Jun 8, 2026
- SHA
- 2565b786
Model card
From the Hugging Face model README
melo-tts-rknn
Chinese/English TTS (Text-to-Speech) — MeloTTS converted for RK3588 NPU.
Models
RKNN checkpoints
| File | Description |
|---|---|
checkpoint/rknn/bert_lml_model.rknn | BERT language model |
checkpoint/rknn/prior_model.rknn | Prior model |
checkpoint/rknn/flow_model.rknn | Flow model |
checkpoint/rknn/decoder_frame31.rknn | Decoder (frame 31) |
checkpoint/rknn/configuration.json | Model configuration |
checkpoint/rknn/tokenizer.json | Tokenizer |
checkpoint/rknn/vocab.txt | Vocabulary |
Text resources
| File | Description |
|---|---|
model/MeloTTS-ONNX/melo_onnx/text/opencpop-strict.txt | Phoneme mapping |
model/MeloTTS-ONNX/melo_onnx/text/tone_sandhi.py | Tone sandhi rules |
Third-party data
| Path | Description |
|---|---|
third_party_data/jieba/dict.txt | Jieba segmentation dictionary |
third_party_data/pypinyin/pinyin_dict.json | Pinyin mapping |
third_party_data/pypinyin/phrases_dict.json | Phrase-level pinyin |
Runtime
Used by the buddy TTS pipeline on RK3588 NPU.
- Created
- Jun 8, 2026
- Updated
- Jun 8, 2026