ipsilondev/voxcpm-spanish-onnx-18k
By ipsilondev · ↓ 0 · ♥ 0 · Updated 2026-01-28
What's in this repo
This page lists the files in ipsilondev/voxcpm-spanish-onnx-18k, including audio_decoder.onnx, audio_encoder.onnx, base_lm_prefill.onnx and more.
- 30 ONNX models (.onnx files)
- 64 files, including 30 .onnx, 14 .data, 10 .json, 7 .py, 2 .md
- Last updated 2026-01-28
ONNX files
| File | Size | Download |
|---|---|---|
| fp32_v2/audio_decoder.onnx | — | Download |
| fp32_v2/audio_encoder.onnx | — | Download |
| fp32_v2/base_lm_prefill.onnx | — | Download |
| fp32_v2/base_lm_step.onnx | — | Download |
| fp32_v2/dit_estimator.onnx | — | Download |
| fp32_v2/local_encoder.onnx | — | Download |
| fp32_v2/projections.onnx | — | Download |
| fp32_v2/residual_lm_prefill.onnx | — | Download |
| fp32_v2/residual_lm_step.onnx | — | Download |
| fp32_v2/stop_head.onnx | — | Download |
Showing 10 of 30 .onnx files — see the full file list on the right.
How to use
These ONNX files can be loaded with ONNX Runtime. To inspect a model graph, upload the file to the Netron viewer. Download links point to the original files on Hugging Face.
README
license: apache-2.0 tags:
- text-to-speech
- onnx
- spanish
- quantized language:
- es
VoxCPM Spanish ONNX Models (18k Steps)
This repository contains ONNX exported models for the VoxCPM Spanish fine-tune (18k steps).
Contents
- fp32_v2/: Full precision FP32 models (Best quality)
- int8_v2/: Quantized INT8 models (Fastest, suitable for mobile/CPU)
- scripts/: Inference scripts for both versions
Usage
FP32 Inference
python scripts/test_onnx_inference_v2.py --onnx-dir fp32_v2 --text "Hola mundo"
INT8 Inference
python scripts/test_onnx_inference_int8.py --onnx-dir int8_v2 --text "Hola mundo"
Performance (CPU)
- FP32: ~56x RTF
- INT8: ~52x RTF (Similar quality to FP32, much smaller size)
Model info
onnx
text-to-speech
spanish
quantized
es
license:apache-2.0
region:us
↓ 0
♥ 0
Created: 2026-01-28
Updated: 2026-01-28
View on Hugging FaceFiles (64)
.gitattributes
DAKEQQ_EXPORT_README.md
README.md
fp32_v2/audio_decoder.onnx
ONNX
fp32_v2/audio_decoder.onnx.data
fp32_v2/audio_encoder.onnx
ONNX
fp32_v2/audio_encoder.onnx.data
fp32_v2/base_lm_prefill.onnx
ONNX
fp32_v2/base_lm_prefill.onnx.data
fp32_v2/base_lm_step.onnx
ONNX
fp32_v2/base_lm_step.onnx.data
fp32_v2/config.json
fp32_v2/dit_estimator.onnx
ONNX
fp32_v2/dit_estimator.onnx.data
fp32_v2/export_info.json
fp32_v2/local_encoder.onnx
ONNX
fp32_v2/local_encoder.onnx.data
fp32_v2/projections.onnx
ONNX
fp32_v2/projections.onnx.data
fp32_v2/residual_lm_prefill.onnx
ONNX
fp32_v2/residual_lm_prefill.onnx.data
fp32_v2/residual_lm_step.onnx
ONNX
fp32_v2/residual_lm_step.onnx.data
fp32_v2/special_tokens_map.json
fp32_v2/stop_head.onnx
ONNX
fp32_v2/stop_head.onnx.data
fp32_v2/text_embedder.onnx
ONNX
fp32_v2/text_embedder.onnx.data
fp32_v2/tokenizer.json
fp32_v2/tokenizer_config.json
int8/VoxCPM_Concat.onnx
ONNX
int8/VoxCPM_Feat_Cond.onnx
ONNX
int8/VoxCPM_Feat_Decoder.onnx
ONNX
int8/VoxCPM_Feat_Encoder.onnx
ONNX
int8/VoxCPM_Main.onnx
ONNX
int8/VoxCPM_Text_Embed.onnx
ONNX
int8/VoxCPM_VAE_Decoder.onnx
ONNX
int8/VoxCPM_VAE_Encoder.onnx
ONNX
int8_v2/audio_decoder.onnx
ONNX
int8_v2/audio_encoder.onnx
ONNX
int8_v2/base_lm_prefill.onnx
ONNX
int8_v2/base_lm_step.onnx
ONNX
int8_v2/base_lm_step.onnx.data
int8_v2/config.json
int8_v2/dit_estimator.onnx
ONNX
int8_v2/export_info.json
int8_v2/local_encoder.onnx
ONNX
int8_v2/projections.onnx
ONNX
int8_v2/residual_lm_prefill.onnx
ONNX
int8_v2/residual_lm_step.onnx
ONNX
int8_v2/residual_lm_step.onnx.data
int8_v2/special_tokens_map.json
int8_v2/stop_head.onnx
ONNX
int8_v2/stop_head.onnx.data
int8_v2/text_embedder.onnx
ONNX
int8_v2/tokenizer.json
int8_v2/tokenizer_config.json
scripts/export_dakeqq_style.py
scripts/optimize_dakeqq_style.py
scripts/test_english_with_spanish_model.py
scripts/test_maru_voice_bilingual.py
scripts/test_onnx_inference_int8.py
scripts/test_onnx_inference_v2.py
scripts/test_spanish_dakeqq_inference.py