Downloads · 30 days
0
tintitu/speech_paraformer-large-contextual_asr_nat-zh-cn-16k-common-vocab8404-onnx
speech_paraformer-large-contextual_asr_nat-zh-cn-16k-common-vocab8404-onnx is a machine learning model from tintitu. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
面向中文自动语音识别(ASR)的端到端非自回归大模型(集成上下文热词偏置网络 Contextual Biasing Network),当前目录包含 ONNX 量化主模型、热词偏置嵌入网络、分词词典、分词词表及解码配置。
Downloads · 30 days
0
Access
Public
Updated Sep 11, 2026
Repo size
834 MB
Likes
0
Public
Click a slice to open those files.
.zip834 MB · 100%
From the Hugging Face model README
面向中文自动语音识别(ASR)的端到端非自回归大模型(集成上下文热词偏置网络 Contextual Biasing Network),当前目录包含 ONNX 量化主模型、热词偏置嵌入网络、分词词典、分词词表及解码配置。
model_quant.onnx:经过 INT8 静态量化的声学主干推理模型model_eb.onnx:热词嵌入偏置网络(Embedding Biaser)seg_dict:热词切词与分词辅助字典tokens.json:字符与词元 Token 映射词表config.yaml:模型推理与热词协同解码配置am.mvn:声学特征 CMVN(倒谱均值方差归一化)统计文件speech_fsmn_vad_zh-cn-16k-common-onnx 进行语音活动检测与长音频断句,滤除环境静音并提高识别效率。punc_ct-transformer_zh-cn-common-vocab272727-onnx)完成自动标点补充。alibaba-damo-academy/FunASRdamo/speech_paraformer-large-contextual_asr_nat-zh-cn-16k-common-vocab8404-onnx本目录包含的是上游模型导出并量化后的 ONNX 推理版本,版权归原作者及阿里巴巴达摩院/通义实验室所有。
遵循上游 FunASR 与 ModelScope 开源协议(Apache-2.0 / ModelScope 社区许可)。商业集成或二次分发时,请保留上游版权声明、许可说明及原始出处链接。
下载后请核对文件大小和 SHA256 校验和。
| 文件名 | 大小 (字节) | SHA-256 校验和 |
|---|---|---|
am.mvn | 11,203 | 29b3c740a2c0cfc6b308126d31d7f265fa2be74f3bb095cd2f143ea970896ae5 |
config.yaml | 2,532 | 1d9057edeaba9e131cb98f26011606497cf3af187d8943525ddb5ee36c836b1b |
model_eb.onnx | 25,618,359 | d31446a5af664291a2922cca253a4200a523f347d6fc3cb1bff356bf60a116b6 |
model_quant.onnx | 871,251,660 | f404e6eb532b54fd95761e2b4be4ed1998e8cff3cb3b930a9bee1f2d556e5035 |
seg_dict | 8,287,834 | 59a2ef803a3f1648ad03a2e1480db1c1ee0c0d7dc4ef4dbd16cea33944329022 |
tokens.json | 93,676 | 2b20c2b12572d682afff84ce1c8d560f67b8b32a4c1f21567411d141ed352127 |
An end-to-end non-autoregressive speech recognition model with contextual hotword biasing for Mandarin Chinese ASR. This directory contains quantized ONNX weights, embedding biaser network, segmentation dictionary, vocabulary mappings, and configuration files.
model_quant.onnx: INT8 statically quantized acoustic inference backbonemodel_eb.onnx: Embedding biaser network for contextual hotword injectionseg_dict: Word segmentation dictionary for hotword tokenizationtokens.json: Character and token mapping dictionaryconfig.yaml: Model configuration and biaser decoding parametersam.mvn: Cepstral mean and variance normalization (CMVN) statisticsspeech_fsmn_vad_zh-cn-16k-common-onnx for voice activity detection and sentence segmentation.punc_ct-transformer_zh-cn-common-vocab272727-onnx).alibaba-damo-academy/FunASRdamo/speech_paraformer-large-contextual_asr_nat-zh-cn-16k-common-vocab8404-onnxThis directory provides a quantized ONNX inference distribution derived from upstream checkpoints. All copyrights belong to the original authors and Alibaba DAMO Academy / Tongyi Lab.
Governed by the upstream FunASR and ModelScope licensing terms (Apache-2.0 / ModelScope Community License). Please retain upstream copyright statements, notices, and source links in downstream redistributions.
Verify file sizes and SHA-256 checksums after downloading.
| Filename | Size (Bytes) | SHA-256 Checksum |
|---|---|---|
am.mvn | 11,203 | 29b3c740a2c0cfc6b308126d31d7f265fa2be74f3bb095cd2f143ea970896ae5 |
config.yaml | 2,532 | 1d9057edeaba9e131cb98f26011606497cf3af187d8943525ddb5ee36c836b1b |
model_eb.onnx | 25,618,359 | d31446a5af664291a2922cca253a4200a523f347d6fc3cb1bff356bf60a116b6 |
model_quant.onnx | 871,251,660 | f404e6eb532b54fd95761e2b4be4ed1998e8cff3cb3b930a9bee1f2d556e5035 |
seg_dict | 8,287,834 | 59a2ef803a3f1648ad03a2e1480db1c1ee0c0d7dc4ef4dbd16cea33944329022 |
tokens.json | 93,676 | 2b20c2b12572d682afff84ce1c8d560f67b8b32a4c1f21567411d141ed352127 |