Commit ·
c005e9b
0
Parent(s):
Duplicate from microsoft/VibeVoice-ASR-HF
Browse filesCo-authored-by: Furu Wei <frontierai@users.noreply.huggingface.co>
- .gitattributes +38 -0
- README.md +545 -0
- chat_template.jinja +26 -0
- config.json +134 -0
- figures/DER.jpg +0 -0
- figures/VibeVoice_ASR_archi.png +3 -0
- figures/cpWER.jpg +0 -0
- figures/language_distribution_horizontal.png +3 -0
- figures/tcpWER.jpg +0 -0
- generation_config.json +12 -0
- model-00001-of-00008.safetensors +3 -0
- model-00002-of-00008.safetensors +3 -0
- model-00003-of-00008.safetensors +3 -0
- model-00004-of-00008.safetensors +3 -0
- model-00005-of-00008.safetensors +3 -0
- model-00006-of-00008.safetensors +3 -0
- model-00007-of-00008.safetensors +3 -0
- model-00008-of-00008.safetensors +3 -0
- model.safetensors.index.json +909 -0
- processor_config.json +18 -0
- tokenizer.json +3 -0
- tokenizer_config.json +30 -0
.gitattributes
ADDED
|
@@ -0,0 +1,38 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
*.7z filter=lfs diff=lfs merge=lfs -text
|
| 2 |
+
*.arrow filter=lfs diff=lfs merge=lfs -text
|
| 3 |
+
*.bin filter=lfs diff=lfs merge=lfs -text
|
| 4 |
+
*.bz2 filter=lfs diff=lfs merge=lfs -text
|
| 5 |
+
*.ckpt filter=lfs diff=lfs merge=lfs -text
|
| 6 |
+
*.ftz filter=lfs diff=lfs merge=lfs -text
|
| 7 |
+
*.gz filter=lfs diff=lfs merge=lfs -text
|
| 8 |
+
*.h5 filter=lfs diff=lfs merge=lfs -text
|
| 9 |
+
*.joblib filter=lfs diff=lfs merge=lfs -text
|
| 10 |
+
*.lfs.* filter=lfs diff=lfs merge=lfs -text
|
| 11 |
+
*.mlmodel filter=lfs diff=lfs merge=lfs -text
|
| 12 |
+
*.model filter=lfs diff=lfs merge=lfs -text
|
| 13 |
+
*.msgpack filter=lfs diff=lfs merge=lfs -text
|
| 14 |
+
*.npy filter=lfs diff=lfs merge=lfs -text
|
| 15 |
+
*.npz filter=lfs diff=lfs merge=lfs -text
|
| 16 |
+
*.onnx filter=lfs diff=lfs merge=lfs -text
|
| 17 |
+
*.ot filter=lfs diff=lfs merge=lfs -text
|
| 18 |
+
*.parquet filter=lfs diff=lfs merge=lfs -text
|
| 19 |
+
*.pb filter=lfs diff=lfs merge=lfs -text
|
| 20 |
+
*.pickle filter=lfs diff=lfs merge=lfs -text
|
| 21 |
+
*.pkl filter=lfs diff=lfs merge=lfs -text
|
| 22 |
+
*.pt filter=lfs diff=lfs merge=lfs -text
|
| 23 |
+
*.pth filter=lfs diff=lfs merge=lfs -text
|
| 24 |
+
*.rar filter=lfs diff=lfs merge=lfs -text
|
| 25 |
+
*.safetensors filter=lfs diff=lfs merge=lfs -text
|
| 26 |
+
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
| 27 |
+
*.tar.* filter=lfs diff=lfs merge=lfs -text
|
| 28 |
+
*.tar filter=lfs diff=lfs merge=lfs -text
|
| 29 |
+
*.tflite filter=lfs diff=lfs merge=lfs -text
|
| 30 |
+
*.tgz filter=lfs diff=lfs merge=lfs -text
|
| 31 |
+
*.wasm filter=lfs diff=lfs merge=lfs -text
|
| 32 |
+
*.xz filter=lfs diff=lfs merge=lfs -text
|
| 33 |
+
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
+
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
+
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
figures/language_distribution_horizontal.png filter=lfs diff=lfs merge=lfs -text
|
| 38 |
+
figures/VibeVoice_ASR_archi.png filter=lfs diff=lfs merge=lfs -text
|
README.md
ADDED
|
@@ -0,0 +1,545 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language:
|
| 3 |
+
- en
|
| 4 |
+
- zh
|
| 5 |
+
- es
|
| 6 |
+
- pt
|
| 7 |
+
- de
|
| 8 |
+
- ja
|
| 9 |
+
- ko
|
| 10 |
+
- fr
|
| 11 |
+
- ru
|
| 12 |
+
- id
|
| 13 |
+
- sv
|
| 14 |
+
- it
|
| 15 |
+
- he
|
| 16 |
+
- nl
|
| 17 |
+
- pl
|
| 18 |
+
- 'no'
|
| 19 |
+
- tr
|
| 20 |
+
- th
|
| 21 |
+
- ar
|
| 22 |
+
- hu
|
| 23 |
+
- ca
|
| 24 |
+
- cs
|
| 25 |
+
- da
|
| 26 |
+
- fa
|
| 27 |
+
- af
|
| 28 |
+
- hi
|
| 29 |
+
- fi
|
| 30 |
+
- et
|
| 31 |
+
- aa
|
| 32 |
+
- el
|
| 33 |
+
- ro
|
| 34 |
+
- vi
|
| 35 |
+
- bg
|
| 36 |
+
- is
|
| 37 |
+
- sl
|
| 38 |
+
- sk
|
| 39 |
+
- lt
|
| 40 |
+
- sw
|
| 41 |
+
- uk
|
| 42 |
+
- kl
|
| 43 |
+
- lv
|
| 44 |
+
- hr
|
| 45 |
+
- ne
|
| 46 |
+
- sr
|
| 47 |
+
- tl
|
| 48 |
+
- yi
|
| 49 |
+
- ms
|
| 50 |
+
- ur
|
| 51 |
+
- mn
|
| 52 |
+
- hy
|
| 53 |
+
- jv
|
| 54 |
+
license: mit
|
| 55 |
+
pipeline_tag: audio-text-to-text
|
| 56 |
+
tags:
|
| 57 |
+
- ASR
|
| 58 |
+
- Diarization
|
| 59 |
+
- Speech-to-Text
|
| 60 |
+
- Transcription
|
| 61 |
+
library_name: transformers
|
| 62 |
+
---
|
| 63 |
+
|
| 64 |
+
|
| 65 |
+
## VibeVoice-ASR (Transformers-compatible version)
|
| 66 |
+
[](https://github.com/microsoft/VibeVoice)
|
| 67 |
+
[](https://huggingface.co/docs/microsoft-azure/foundry/guides/vibevoice-asr)
|
| 68 |
+
[](https://aka.ms/vibevoice-asr)
|
| 69 |
+
[](https://arxiv.org/pdf/2601.18184)
|
| 70 |
+
|
| 71 |
+
**VibeVoice-ASR** is a unified speech-to-text model designed to handle **60-minute long-form audio** in a single pass, generating structured transcriptions containing **Who (Speaker), When (Timestamps), and What (Content)**, with support for **Customized Hotwords** and over **50 languages**.
|
| 72 |
+
|
| 73 |
+
➡️ **Demo:** [VibeVoice-ASR-Demo](https://aka.ms/vibevoice-asr)<br>
|
| 74 |
+
➡️ **Report:** [VibeVoice-ASR Technical Report](https://arxiv.org/pdf/2601.18184)<br>
|
| 75 |
+
|
| 76 |
+
<p align="left">
|
| 77 |
+
<img src="figures/VibeVoice_ASR_archi.png" alt="VibeVoice-ASR Architecture" height="250px">
|
| 78 |
+
</p>
|
| 79 |
+
|
| 80 |
+
|
| 81 |
+
## 🔥 Key Features
|
| 82 |
+
|
| 83 |
+
|
| 84 |
+
- **🕒 60-minute Single-Pass Processing**:
|
| 85 |
+
Unlike conventional ASR models that slice audio into short chunks (often losing global context), VibeVoice ASR accepts up to **60 minutes** of continuous audio input within 64K token length. This ensures consistent speaker tracking and semantic coherence across the entire hour.
|
| 86 |
+
|
| 87 |
+
- **👤 Customized Hotwords**:
|
| 88 |
+
Users can provide customized hotwords (e.g., specific names, technical terms, or background info) to guide the recognition process, significantly improving accuracy on domain-specific content.
|
| 89 |
+
|
| 90 |
+
- **📝 Rich Transcription (Who, When, What)**:
|
| 91 |
+
The model jointly performs ASR, diarization, and timestamping, producing a structured output that indicates *who* said *what* and *when*.
|
| 92 |
+
|
| 93 |
+
- **🌍 Multilingual & Code-Switching Support**:
|
| 94 |
+
It supports over 50 languages, requires no explicit language setting, and natively handles code-switching within and across utterances. Language distribution can be found [here](#language-distribution).
|
| 95 |
+
|
| 96 |
+
|
| 97 |
+
## Usage
|
| 98 |
+
|
| 99 |
+
### Setup
|
| 100 |
+
|
| 101 |
+
VibeVoice-ASR is available as of v5.3.0 of Transformers!
|
| 102 |
+
|
| 103 |
+
```
|
| 104 |
+
pip install "transformers>=5.3.0"
|
| 105 |
+
```
|
| 106 |
+
|
| 107 |
+
### Loading model
|
| 108 |
+
|
| 109 |
+
```python
|
| 110 |
+
from transformers import AutoProcessor, VibeVoiceAsrForConditionalGeneration
|
| 111 |
+
|
| 112 |
+
model_id = "microsoft/VibeVoice-ASR-HF"
|
| 113 |
+
processor = AutoProcessor.from_pretrained(model_id)
|
| 114 |
+
model = VibeVoiceAsrForConditionalGeneration.from_pretrained(model_id)
|
| 115 |
+
```
|
| 116 |
+
|
| 117 |
+
### Speaker-timestamped transcription
|
| 118 |
+
|
| 119 |
+
A notable feature of VibeVoice ASR is its ability to transcribe multi-speaker content, denoting who spoke and when.
|
| 120 |
+
|
| 121 |
+
The example below transcribes the following audio.
|
| 122 |
+
|
| 123 |
+
<audio controls>
|
| 124 |
+
<source src="https://huggingface.co/datasets/bezzam/vibevoice_samples/resolve/main/example_output/VibeVoice-1.5B_output.wav" type="audio/wav">
|
| 125 |
+
</audio>
|
| 126 |
+
|
| 127 |
+
```python
|
| 128 |
+
from transformers import AutoProcessor, VibeVoiceAsrForConditionalGeneration
|
| 129 |
+
|
| 130 |
+
model_id = "microsoft/VibeVoice-ASR-HF"
|
| 131 |
+
processor = AutoProcessor.from_pretrained(model_id)
|
| 132 |
+
model = VibeVoiceAsrForConditionalGeneration.from_pretrained(model_id, device_map="auto")
|
| 133 |
+
print(f"Model loaded on {model.device} with dtype {model.dtype}")
|
| 134 |
+
|
| 135 |
+
# Prepare inputs using `apply_transcription_request`
|
| 136 |
+
inputs = processor.apply_transcription_request(
|
| 137 |
+
audio="https://huggingface.co/datasets/bezzam/vibevoice_samples/resolve/main/example_output/VibeVoice-1.5B_output.wav",
|
| 138 |
+
).to(model.device, model.dtype)
|
| 139 |
+
|
| 140 |
+
# Apply model
|
| 141 |
+
output_ids = model.generate(**inputs)
|
| 142 |
+
generated_ids = output_ids[:, inputs["input_ids"].shape[1] :]
|
| 143 |
+
transcription = processor.decode(generated_ids)[0]
|
| 144 |
+
print("\n" + "=" * 60)
|
| 145 |
+
print("RAW OUTPUT")
|
| 146 |
+
print("=" * 60)
|
| 147 |
+
print(transcription)
|
| 148 |
+
|
| 149 |
+
transcription = processor.decode(generated_ids, return_format="parsed")[0]
|
| 150 |
+
print("\n" + "=" * 60)
|
| 151 |
+
print("TRANSCRIPTION (list of dicts)")
|
| 152 |
+
print("=" * 60)
|
| 153 |
+
for speaker_transcription in transcription:
|
| 154 |
+
print(speaker_transcription)
|
| 155 |
+
|
| 156 |
+
# Remove speaker labels, only get raw transcription
|
| 157 |
+
transcription = processor.decode(generated_ids, return_format="transcription_only")[0]
|
| 158 |
+
print("\n" + "=" * 60)
|
| 159 |
+
print("TRANSCRIPTION ONLY")
|
| 160 |
+
print("=" * 60)
|
| 161 |
+
print(transcription)
|
| 162 |
+
|
| 163 |
+
"""
|
| 164 |
+
============================================================
|
| 165 |
+
RAW OUTPUT
|
| 166 |
+
============================================================
|
| 167 |
+
<|im_start|>assistant
|
| 168 |
+
[{"Start":0,"End":15.43,"Speaker":0,"Content":"Hello everyone and welcome to the Vibe Voice podcast. I'm your host, Alex, and today we're getting into one of the biggest debates in all of sports: who's the greatest basketball player of all time? I'm so excited to have Sam here to talk about it with me."},{"Start":15.43,"End":21.05,"Speaker":1,"Content":"Thanks so much for having me, Alex. And you're absolutely right. This question always brings out some seriously strong feelings."},{"Start":21.05,"End":31.66,"Speaker":0,"Content":"Okay, so let's get right into it. For me, it has to be Michael Jordan. Six trips to the finals, six championships. That kind of perfection is just incredible."},{"Start":31.66,"End":40.93,"Speaker":1,"Content":"Oh man, the first thing that always pops into my head is that shot against the Cleveland Cavaliers back in '89. Jordan just rises, hangs in the air forever, and just sinks it."}]<|im_end|>
|
| 169 |
+
<|endoftext|>
|
| 170 |
+
|
| 171 |
+
============================================================
|
| 172 |
+
TRANSCRIPTION (list of dicts)
|
| 173 |
+
============================================================
|
| 174 |
+
{'Start': 0, 'End': 15.43, 'Speaker': 0, 'Content': "Hello everyone and welcome to the Vibe Voice podcast. I'm your host, Alex, and today we're getting into one of the biggest debates in all of sports: who's the greatest basketball player of all time? I'm so excited to have Sam here to talk about it with me."}
|
| 175 |
+
{'Start': 15.43, 'End': 21.05, 'Speaker': 1, 'Content': "Thanks so much for having me, Alex. And you're absolutely right. This question always brings out some seriously strong feelings."}
|
| 176 |
+
{'Start': 21.05, 'End': 31.66, 'Speaker': 0, 'Content': "Okay, so let's get right into it. For me, it has to be Michael Jordan. Six trips to the finals, six championships. That kind of perfection is just incredible."}
|
| 177 |
+
{'Start': 31.66, 'End': 40.93, 'Speaker': 1, 'Content': "Oh man, the first thing that always pops into my head is that shot against the Cleveland Cavaliers back in '89. Jordan just rises, hangs in the air forever, and just sinks it."}
|
| 178 |
+
|
| 179 |
+
============================================================
|
| 180 |
+
TRANSCRIPTION ONLY
|
| 181 |
+
============================================================
|
| 182 |
+
Hello everyone and welcome to the Vibe Voice podcast. I'm your host, Alex, and today we're getting into one of the biggest debates in all of sports: who's the greatest basketball player of all time? I'm so excited to have Sam here to talk about it with me. Thanks so much for having me, Alex. And you're absolutely right. This question always brings out some seriously strong feelings. Okay, so let's get right into it. For me, it has to be Michael Jordan. Six trips to the finals, six championships. That kind of perfection is just incredible. Oh man, the first thing that always pops into my head is that shot against the Cleveland Cavaliers back in '89. Jordan just rises, hangs in the air forever, and just sinks it.
|
| 183 |
+
"""
|
| 184 |
+
```
|
| 185 |
+
|
| 186 |
+
The VibeVoice ASR model is trained to generate a string that resembles a JSON structure. The flag `return_format="parsed"` tries to return the generated output as a list of dicts, while `return_format="transcription_only"` tries to extract only the transcribed audio. If they fail, the generated output is returned as-is.
|
| 187 |
+
|
| 188 |
+
### Providing context
|
| 189 |
+
|
| 190 |
+
It is also possible to provide context. This can be useful if certain words cannot be transcribed correctly, such as proper nouns.
|
| 191 |
+
|
| 192 |
+
Below we transcribe an audio where the speaker (with a German accent) talks about VibeVoice, comparing with and without the context "About VibeVoice".
|
| 193 |
+
|
| 194 |
+
<audio controls>
|
| 195 |
+
<source src="https://huggingface.co/datasets/bezzam/vibevoice_samples/resolve/main/realtime_model/vibevoice_tts_german.wav" type="audio/wav">
|
| 196 |
+
</audio>
|
| 197 |
+
|
| 198 |
+
```python
|
| 199 |
+
from transformers import AutoProcessor, VibeVoiceAsrForConditionalGeneration
|
| 200 |
+
|
| 201 |
+
model_id = "microsoft/VibeVoice-ASR-HF"
|
| 202 |
+
processor = AutoProcessor.from_pretrained(model_id)
|
| 203 |
+
model = VibeVoiceAsrForConditionalGeneration.from_pretrained(model_id, device_map="auto")
|
| 204 |
+
print(f"Model loaded on {model.device} with dtype {model.dtype}")
|
| 205 |
+
|
| 206 |
+
# Without context
|
| 207 |
+
inputs = processor.apply_transcription_request(
|
| 208 |
+
audio="https://huggingface.co/datasets/bezzam/vibevoice_samples/resolve/main/realtime_model/vibevoice_tts_german.wav",
|
| 209 |
+
).to(model.device, model.dtype)
|
| 210 |
+
output_ids = model.generate(**inputs)
|
| 211 |
+
generated_ids = output_ids[:, inputs["input_ids"].shape[1] :]
|
| 212 |
+
transcription = processor.decode(generated_ids, return_format="transcription_only")[0]
|
| 213 |
+
print(f"WITHOUT CONTEXT: {transcription}")
|
| 214 |
+
|
| 215 |
+
# With context
|
| 216 |
+
inputs = processor.apply_transcription_request(
|
| 217 |
+
audio="https://huggingface.co/datasets/bezzam/vibevoice_samples/resolve/main/realtime_model/vibevoice_tts_german.wav",
|
| 218 |
+
prompt="About VibeVoice",
|
| 219 |
+
).to(model.device, model.dtype)
|
| 220 |
+
output_ids = model.generate(**inputs)
|
| 221 |
+
generated_ids = output_ids[:, inputs["input_ids"].shape[1] :]
|
| 222 |
+
transcription = processor.decode(generated_ids, return_format="transcription_only")[0]
|
| 223 |
+
print(f"WITH CONTEXT : {transcription}")
|
| 224 |
+
|
| 225 |
+
"""
|
| 226 |
+
WITHOUT CONTEXT: Revevoices is a novel framework designed for generating expressive, long-form, multi-speaker conversational audio.
|
| 227 |
+
WITH CONTEXT : VibeVoice is this novel framework designed for generating expressive, long-form, multi-speaker, conversational audio.
|
| 228 |
+
"""
|
| 229 |
+
```
|
| 230 |
+
|
| 231 |
+
|
| 232 |
+
### Batch inference
|
| 233 |
+
|
| 234 |
+
Batch inference is possible by passing a list of audio and (if provided) a list of prompts of equal length.
|
| 235 |
+
|
| 236 |
+
```python
|
| 237 |
+
from transformers import AutoProcessor, VibeVoiceAsrForConditionalGeneration
|
| 238 |
+
|
| 239 |
+
model_id = "microsoft/VibeVoice-ASR-HF"
|
| 240 |
+
audio = [
|
| 241 |
+
"https://huggingface.co/datasets/bezzam/vibevoice_samples/resolve/main/realtime_model/vibevoice_tts_german.wav",
|
| 242 |
+
"https://huggingface.co/datasets/bezzam/vibevoice_samples/resolve/main/example_output/VibeVoice-1.5B_output.wav"
|
| 243 |
+
]
|
| 244 |
+
prompts = ["About VibeVoice", None]
|
| 245 |
+
|
| 246 |
+
processor = AutoProcessor.from_pretrained(model_id)
|
| 247 |
+
model = VibeVoiceAsrForConditionalGeneration.from_pretrained(model_id, device_map="auto")
|
| 248 |
+
print(f"Model loaded on {model.device} with dtype {model.dtype}")
|
| 249 |
+
|
| 250 |
+
inputs = processor.apply_transcription_request(audio, prompt=prompts).to(model.device, model.dtype)
|
| 251 |
+
output_ids = model.generate(**inputs)
|
| 252 |
+
generated_ids = output_ids[:, inputs["input_ids"].shape[1] :]
|
| 253 |
+
transcription = processor.decode(generated_ids, return_format="transcription_only")
|
| 254 |
+
|
| 255 |
+
print(transcription)
|
| 256 |
+
```
|
| 257 |
+
|
| 258 |
+
### Adjusting tokenizer chunk (e.g. if out-of-memory)
|
| 259 |
+
|
| 260 |
+
A key feature of VibeVoice ASR is that it can transcribe up to 60 minutes of continuous audio. This is done by chunking audio into 60-second segments (1440000 samples at 24kHz) and caching the convolution states between each segment.
|
| 261 |
+
|
| 262 |
+
However, if chunks of 60 seconds are too large for your device, the `tokenizer_chunk_size` argument passed to `generate` can be adjusted. *Note it should be a multiple of the hop length (3200 for the original acoustic tokenizer).*
|
| 263 |
+
|
| 264 |
+
```python
|
| 265 |
+
from transformers import AutoProcessor, VibeVoiceAsrForConditionalGeneration
|
| 266 |
+
|
| 267 |
+
tokenizer_chunk_size = 64000 # default is 1440000 (60s @ 24kHz)
|
| 268 |
+
model_id = "microsoft/VibeVoice-ASR-HF"
|
| 269 |
+
audio = [
|
| 270 |
+
"https://huggingface.co/datasets/bezzam/vibevoice_samples/resolve/main/realtime_model/vibevoice_tts_german.wav",
|
| 271 |
+
"https://huggingface.co/datasets/bezzam/vibevoice_samples/resolve/main/example_output/VibeVoice-1.5B_output.wav"
|
| 272 |
+
]
|
| 273 |
+
prompts = ["About VibeVoice", None]
|
| 274 |
+
|
| 275 |
+
processor = AutoProcessor.from_pretrained(model_id)
|
| 276 |
+
model = VibeVoiceAsrForConditionalGeneration.from_pretrained(model_id, device_map="auto")
|
| 277 |
+
print(f"Model loaded on {model.device} with dtype {model.dtype}")
|
| 278 |
+
|
| 279 |
+
inputs = processor.apply_transcription_request(audio, prompt=prompts).to(model.device, model.dtype)
|
| 280 |
+
output_ids = model.generate(**inputs, tokenizer_chunk_size=tokenizer_chunk_size)
|
| 281 |
+
generated_ids = output_ids[:, inputs["input_ids"].shape[1] :]
|
| 282 |
+
transcription = processor.decode(generated_ids, return_format="transcription_only")
|
| 283 |
+
print(transcription)
|
| 284 |
+
```
|
| 285 |
+
|
| 286 |
+
### Chat template
|
| 287 |
+
|
| 288 |
+
VibeVoice ASR also accepts chat template inputs (`apply_transcription_request` is actually a wrapper for `apply_chat_template` for convenience):
|
| 289 |
+
```python
|
| 290 |
+
from transformers import AutoProcessor, VibeVoiceAsrForConditionalGeneration
|
| 291 |
+
|
| 292 |
+
model_id = "microsoft/VibeVoice-ASR-HF"
|
| 293 |
+
processor = AutoProcessor.from_pretrained(model_id)
|
| 294 |
+
model = VibeVoiceAsrForConditionalGeneration.from_pretrained(model_id, device_map="auto")
|
| 295 |
+
|
| 296 |
+
chat_template = [
|
| 297 |
+
[
|
| 298 |
+
{
|
| 299 |
+
"role": "user",
|
| 300 |
+
"content": [
|
| 301 |
+
{"type": "text", "text": "About VibeVoice"},
|
| 302 |
+
{
|
| 303 |
+
"type": "audio",
|
| 304 |
+
"path": "https://huggingface.co/datasets/bezzam/vibevoice_samples/resolve/main/realtime_model/vibevoice_tts_german.wav",
|
| 305 |
+
},
|
| 306 |
+
],
|
| 307 |
+
}
|
| 308 |
+
],
|
| 309 |
+
[
|
| 310 |
+
{
|
| 311 |
+
"role": "user",
|
| 312 |
+
"content": [
|
| 313 |
+
{
|
| 314 |
+
"type": "audio",
|
| 315 |
+
"path": "https://huggingface.co/datasets/bezzam/vibevoice_samples/resolve/main/example_output/VibeVoice-1.5B_output.wav",
|
| 316 |
+
},
|
| 317 |
+
],
|
| 318 |
+
}
|
| 319 |
+
],
|
| 320 |
+
]
|
| 321 |
+
|
| 322 |
+
inputs = processor.apply_chat_template(
|
| 323 |
+
chat_template,
|
| 324 |
+
tokenize=True,
|
| 325 |
+
return_dict=True,
|
| 326 |
+
).to(model.device, model.dtype)
|
| 327 |
+
|
| 328 |
+
output_ids = model.generate(**inputs)
|
| 329 |
+
generated_ids = output_ids[:, inputs["input_ids"].shape[1] :]
|
| 330 |
+
transcription = processor.decode(generated_ids, return_format="transcription_only")
|
| 331 |
+
print(transcription)
|
| 332 |
+
```
|
| 333 |
+
|
| 334 |
+
### Training
|
| 335 |
+
|
| 336 |
+
VibeVoice ASR can be trained with the loss outputted by the model.
|
| 337 |
+
|
| 338 |
+
```python
|
| 339 |
+
from transformers import AutoProcessor, VibeVoiceAsrForConditionalGeneration
|
| 340 |
+
|
| 341 |
+
model_id = "microsoft/VibeVoice-ASR-HF"
|
| 342 |
+
processor = AutoProcessor.from_pretrained(model_id)
|
| 343 |
+
model = VibeVoiceAsrForConditionalGeneration.from_pretrained(model_id, device_map="auto")
|
| 344 |
+
model.train()
|
| 345 |
+
|
| 346 |
+
# Prepare batch of 2
|
| 347 |
+
# -- NOTE: the original model is trained to output transcription, speaker ID, and timestamps in JSON-like format. Below we are only using the transcription text as the label
|
| 348 |
+
chat_template = [
|
| 349 |
+
[
|
| 350 |
+
{
|
| 351 |
+
"role": "user",
|
| 352 |
+
"content": [
|
| 353 |
+
{"type": "text", "text": "VibeVoice is this novel framework designed for generating expressive, long-form, multi-speaker, conversational audio."},
|
| 354 |
+
{
|
| 355 |
+
"type": "audio",
|
| 356 |
+
"path": "https://huggingface.co/datasets/bezzam/vibevoice_samples/resolve/main/realtime_model/vibevoice_tts_german.wav",
|
| 357 |
+
},
|
| 358 |
+
],
|
| 359 |
+
}
|
| 360 |
+
],
|
| 361 |
+
[
|
| 362 |
+
{
|
| 363 |
+
"role": "user",
|
| 364 |
+
"content": [
|
| 365 |
+
{"type": "text", "text": "Hello everyone and welcome to the VibeVoice podcast. I'm your host, Alex, and today we're getting into one of the biggest debates in all of sports: who's the greatest basketball player of all time? I'm so excited to have Sam here to talk about it with me. Thanks so much for having me, Alex. And you're absolutely right. This question always brings out some seriously strong feelings. Okay, so let's get right into it. For me, it has to be Michael Jordan. Six trips to the finals, six championships. That kind of perfection is just incredible. Oh man, the first thing that always pops into my head is that shot against the Cleveland Cavaliers back in '89. Jordan just rises, hangs in the air forever, and just sinks it."},
|
| 366 |
+
{
|
| 367 |
+
"type": "audio",
|
| 368 |
+
"path": "https://huggingface.co/datasets/bezzam/vibevoice_samples/resolve/main/example_output/VibeVoice-1.5B_output.wav",
|
| 369 |
+
},
|
| 370 |
+
],
|
| 371 |
+
}
|
| 372 |
+
],
|
| 373 |
+
]
|
| 374 |
+
inputs = processor.apply_chat_template(
|
| 375 |
+
chat_template,
|
| 376 |
+
tokenize=True,
|
| 377 |
+
return_dict=True,
|
| 378 |
+
output_labels=True,
|
| 379 |
+
).to(model.device, model.dtype)
|
| 380 |
+
|
| 381 |
+
loss = model(**inputs).loss
|
| 382 |
+
print("Loss:", loss.item())
|
| 383 |
+
loss.backward()
|
| 384 |
+
```
|
| 385 |
+
|
| 386 |
+
### Torch compile
|
| 387 |
+
|
| 388 |
+
The model can be compiled for faster inference/training.
|
| 389 |
+
```python
|
| 390 |
+
import time
|
| 391 |
+
import torch
|
| 392 |
+
from transformers import AutoProcessor, VibeVoiceAsrForConditionalGeneration
|
| 393 |
+
|
| 394 |
+
model_id = "microsoft/VibeVoice-ASR-HF"
|
| 395 |
+
|
| 396 |
+
num_warmup = 5
|
| 397 |
+
num_runs = 20
|
| 398 |
+
|
| 399 |
+
# Load processor + model
|
| 400 |
+
processor = AutoProcessor.from_pretrained(model_id)
|
| 401 |
+
model = VibeVoiceAsrForConditionalGeneration.from_pretrained(model_id, torch_dtype=torch.bfloat16,).to("cuda")
|
| 402 |
+
|
| 403 |
+
# Prepare static inputs
|
| 404 |
+
chat_template = [
|
| 405 |
+
[
|
| 406 |
+
{
|
| 407 |
+
"role": "user",
|
| 408 |
+
"content": [
|
| 409 |
+
{
|
| 410 |
+
"type": "text",
|
| 411 |
+
"text": "VibeVoice is this novel framework designed for generating expressive, long-form, multi-speaker, conversational audio.",
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"type": "audio",
|
| 415 |
+
"path": "https://huggingface.co/datasets/bezzam/vibevoice_samples/resolve/main/realtime_model/vibevoice_tts_german.wav",
|
| 416 |
+
},
|
| 417 |
+
],
|
| 418 |
+
}
|
| 419 |
+
],
|
| 420 |
+
] * 4 # batch size 4
|
| 421 |
+
inputs = processor.apply_chat_template(
|
| 422 |
+
chat_template,
|
| 423 |
+
tokenize=True,
|
| 424 |
+
return_dict=True,
|
| 425 |
+
).to("cuda", torch.bfloat16)
|
| 426 |
+
|
| 427 |
+
# Benchmark without compile
|
| 428 |
+
print("Warming up without compile...")
|
| 429 |
+
with torch.no_grad():
|
| 430 |
+
for _ in range(num_warmup):
|
| 431 |
+
_ = model(**inputs)
|
| 432 |
+
|
| 433 |
+
torch.cuda.synchronize()
|
| 434 |
+
|
| 435 |
+
print("\nBenchmarking without torch.compile...")
|
| 436 |
+
torch.cuda.synchronize()
|
| 437 |
+
start = time.time()
|
| 438 |
+
with torch.no_grad():
|
| 439 |
+
for _ in range(num_runs):
|
| 440 |
+
_ = model(**inputs)
|
| 441 |
+
torch.cuda.synchronize()
|
| 442 |
+
no_compile_time = (time.time() - start) / num_runs
|
| 443 |
+
print(f"Average time without compile: {no_compile_time:.4f}s")
|
| 444 |
+
|
| 445 |
+
# Benchmark with compile
|
| 446 |
+
print("\nCompiling model...")
|
| 447 |
+
model = torch.compile(model)
|
| 448 |
+
|
| 449 |
+
print("Warming up with compile (includes graph capture)...")
|
| 450 |
+
with torch.no_grad():
|
| 451 |
+
for _ in range(num_warmup):
|
| 452 |
+
_ = model(**inputs)
|
| 453 |
+
|
| 454 |
+
torch.cuda.synchronize()
|
| 455 |
+
|
| 456 |
+
print("\nBenchmarking with torch.compile...")
|
| 457 |
+
torch.cuda.synchronize()
|
| 458 |
+
start = time.time()
|
| 459 |
+
with torch.no_grad():
|
| 460 |
+
for _ in range(num_runs):
|
| 461 |
+
_ = model(**inputs)
|
| 462 |
+
torch.cuda.synchronize()
|
| 463 |
+
compile_time = (time.time() - start) / num_runs
|
| 464 |
+
print(f"Average time with compile: {compile_time:.4f}s")
|
| 465 |
+
|
| 466 |
+
speedup = no_compile_time / compile_time
|
| 467 |
+
print(f"\nSpeedup: {speedup:.2f}x")
|
| 468 |
+
```
|
| 469 |
+
|
| 470 |
+
### Pipeline usage
|
| 471 |
+
|
| 472 |
+
The model can be used as a pipeline, but you will have to define your own methods for parsing the raw output.
|
| 473 |
+
|
| 474 |
+
```python
|
| 475 |
+
from transformers import pipeline
|
| 476 |
+
|
| 477 |
+
model_id = "microsoft/VibeVoice-ASR-HF"
|
| 478 |
+
pipe = pipeline("any-to-any", model=model_id, device_map="auto")
|
| 479 |
+
chat_template = [
|
| 480 |
+
{
|
| 481 |
+
"role": "user",
|
| 482 |
+
"content": [
|
| 483 |
+
{"type": "text", "text": "About VibeVoice"},
|
| 484 |
+
{
|
| 485 |
+
"type": "audio",
|
| 486 |
+
"path": "https://huggingface.co/datasets/bezzam/vibevoice_samples/resolve/main/realtime_model/vibevoice_tts_german.wav",
|
| 487 |
+
},
|
| 488 |
+
],
|
| 489 |
+
}
|
| 490 |
+
]
|
| 491 |
+
outputs = pipe(text=chat_template, return_full_text=False)
|
| 492 |
+
|
| 493 |
+
print("\n" + "=" * 60)
|
| 494 |
+
print("RAW PIPELINE OUTPUT")
|
| 495 |
+
print("=" * 60)
|
| 496 |
+
print(outputs)
|
| 497 |
+
|
| 498 |
+
"""
|
| 499 |
+
============================================================
|
| 500 |
+
RAW PIPELINE OUTPUT
|
| 501 |
+
============================================================
|
| 502 |
+
[{'input_text': [{'role': 'user', 'content': [{'type': 'text', 'text': 'About VibeVoice'}, {'type': 'audio', 'path': 'https://huggingface.co/datasets/bezzam/vibevoice_samples/resolve/main/realtime_model/vibevoice_tts_german.wav'}]}], 'generated_text': 'assistant\n[{"Start":0.0,"End":7.56,"Speaker":0,"Content":"VibeVoice is this novel framework designed for generating expressive, long-form, multi-speaker conversational audio."}]\n'}]
|
| 503 |
+
"""
|
| 504 |
+
```
|
| 505 |
+
|
| 506 |
+
## Evaluation
|
| 507 |
+
|
| 508 |
+
Below are results from the [technical report](https://arxiv.org/pdf/2601.18184).
|
| 509 |
+
|
| 510 |
+
<p align="center">
|
| 511 |
+
<img src="figures/DER.jpg" alt="DER" width="70%">
|
| 512 |
+
<img src="figures/cpWER.jpg" alt="cpWER" width="70%">
|
| 513 |
+
<img src="figures/tcpWER.jpg" alt="tcpWER" width="70%">
|
| 514 |
+
</p>
|
| 515 |
+
|
| 516 |
+
|
| 517 |
+
### Open ASR Leaderboard
|
| 518 |
+
|
| 519 |
+
On the [Open ASR leaderboard](https://huggingface.co/spaces/hf-audio/open_asr_leaderboard), the following results were obtained:
|
| 520 |
+
|
| 521 |
+
| Dataset | WER (%) |
|
| 522 |
+
| ---------------------- | -------- |
|
| 523 |
+
| ami_test | 17.20 |
|
| 524 |
+
| earnings22_test | 13.17 |
|
| 525 |
+
| gigaspeech_test | 9.67 |
|
| 526 |
+
| librispeech_test.clean | 2.20 |
|
| 527 |
+
| librispeech_test.other | 5.51 |
|
| 528 |
+
| spgispeech_test | 3.80 |
|
| 529 |
+
| tedlium_test | 2.57 |
|
| 530 |
+
| voxpopuli_test | 8.01 |
|
| 531 |
+
| **Average** | **7.77** |
|
| 532 |
+
| **RTFx** | **51.80** |
|
| 533 |
+
|
| 534 |
+
## Language Distribution
|
| 535 |
+
<p align="center">
|
| 536 |
+
<img src="figures/language_distribution_horizontal.png" alt="Language Distribution" width="80%">
|
| 537 |
+
</p>
|
| 538 |
+
|
| 539 |
+
|
| 540 |
+
## License
|
| 541 |
+
This project is licensed under the MIT License.
|
| 542 |
+
|
| 543 |
+
## Contact
|
| 544 |
+
This project was conducted by members of Microsoft Research. We welcome feedback and collaboration from our audience. If you have suggestions, questions, or observe unexpected/offensive behavior in our technology, please contact us at VibeVoice@microsoft.com.
|
| 545 |
+
If the team receives reports of undesired behavior or identifies issues independently, we will update this repository with appropriate mitigations.
|
chat_template.jinja
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{%- set system_prompt = system_prompt | default("You are a helpful assistant that transcribes audio input into text output in JSON format.") -%}
|
| 2 |
+
<|im_start|>system
|
| 3 |
+
{{ system_prompt }}<|im_end|>
|
| 4 |
+
{%- set audio_token = audio_token | default("<|box_start|>") -%}
|
| 5 |
+
{%- set audio_start_token = "<|object_ref_start|>" -%}
|
| 6 |
+
{%- set audio_end_token = "<|object_ref_end|>" -%}
|
| 7 |
+
{%- for message in messages -%}
|
| 8 |
+
{%- if message['role'] == 'user' -%}
|
| 9 |
+
{{ '
|
| 10 |
+
' }}<|im_start|>user{{ '
|
| 11 |
+
' }}{%- set text_items = message['content'] | selectattr('type', 'equalto', 'text') | list -%}
|
| 12 |
+
{%- set context_text = text_items[0]['text'] if text_items else none -%}
|
| 13 |
+
{%- for item in message['content'] -%}
|
| 14 |
+
{%- if item['type'] == 'audio' -%}
|
| 15 |
+
{{ audio_start_token }}{{ audio_token }}{{ audio_end_token }}{{ "
|
| 16 |
+
" }}{%- if context_text -%}
|
| 17 |
+
This is a <|AUDIO_DURATION|> seconds audio, with extra info: {{ context_text }}
|
| 18 |
+
|
| 19 |
+
Please transcribe it with these keys: Start time, End time, Speaker ID, Content{%- else -%}
|
| 20 |
+
This is a <|AUDIO_DURATION|> seconds audio, please transcribe it with these keys: Start time, End time, Speaker ID, Content{%- endif -%}
|
| 21 |
+
{%- endif -%}
|
| 22 |
+
{%- endfor -%}
|
| 23 |
+
<|im_end|>{{ '
|
| 24 |
+
' }}
|
| 25 |
+
{%- endif -%}
|
| 26 |
+
{%- endfor -%}
|
config.json
ADDED
|
@@ -0,0 +1,134 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"acoustic_tokenizer_chunk_size": 1440000,
|
| 3 |
+
"acoustic_tokenizer_encoder_config": {
|
| 4 |
+
"channels": 1,
|
| 5 |
+
"depths": [
|
| 6 |
+
3,
|
| 7 |
+
3,
|
| 8 |
+
3,
|
| 9 |
+
3,
|
| 10 |
+
3,
|
| 11 |
+
3,
|
| 12 |
+
8
|
| 13 |
+
],
|
| 14 |
+
"downsampling_ratios": [
|
| 15 |
+
2,
|
| 16 |
+
2,
|
| 17 |
+
4,
|
| 18 |
+
5,
|
| 19 |
+
5,
|
| 20 |
+
8
|
| 21 |
+
],
|
| 22 |
+
"dtype": "bfloat16",
|
| 23 |
+
"ffn_expansion": 4,
|
| 24 |
+
"hidden_act": "gelu",
|
| 25 |
+
"hidden_size": 64,
|
| 26 |
+
"initializer_range": 0.01,
|
| 27 |
+
"kernel_size": 7,
|
| 28 |
+
"layer_scale_init_value": 1e-06,
|
| 29 |
+
"model_type": "vibevoice_acoustic_tokenizer_encoder",
|
| 30 |
+
"num_filters": 32,
|
| 31 |
+
"rms_norm_eps": 1e-05,
|
| 32 |
+
"vae_std": 0.625,
|
| 33 |
+
"weight_init_value": 0.01
|
| 34 |
+
},
|
| 35 |
+
"architectures": [
|
| 36 |
+
"VibeVoiceAsrForConditionalGeneration"
|
| 37 |
+
],
|
| 38 |
+
"audio_bos_token_id": 151646,
|
| 39 |
+
"audio_eos_token_id": 151647,
|
| 40 |
+
"audio_token_id": 151648,
|
| 41 |
+
"dtype": "bfloat16",
|
| 42 |
+
"model_type": "vibevoice_asr",
|
| 43 |
+
"semantic_tokenizer_encoder_config": {
|
| 44 |
+
"channels": 1,
|
| 45 |
+
"depths": [
|
| 46 |
+
3,
|
| 47 |
+
3,
|
| 48 |
+
3,
|
| 49 |
+
3,
|
| 50 |
+
3,
|
| 51 |
+
3,
|
| 52 |
+
8
|
| 53 |
+
],
|
| 54 |
+
"downsampling_ratios": [
|
| 55 |
+
2,
|
| 56 |
+
2,
|
| 57 |
+
4,
|
| 58 |
+
5,
|
| 59 |
+
5,
|
| 60 |
+
8
|
| 61 |
+
],
|
| 62 |
+
"dtype": "bfloat16",
|
| 63 |
+
"ffn_expansion": 4,
|
| 64 |
+
"hidden_act": "gelu",
|
| 65 |
+
"hidden_size": 128,
|
| 66 |
+
"initializer_range": 0.01,
|
| 67 |
+
"kernel_size": 7,
|
| 68 |
+
"layer_scale_init_value": 1e-06,
|
| 69 |
+
"model_type": "vibevoice_acoustic_tokenizer_encoder",
|
| 70 |
+
"num_filters": 32,
|
| 71 |
+
"rms_norm_eps": 1e-05,
|
| 72 |
+
"vae_std": 0.625,
|
| 73 |
+
"weight_init_value": 0.01
|
| 74 |
+
},
|
| 75 |
+
"text_config": {
|
| 76 |
+
"attention_dropout": 0.0,
|
| 77 |
+
"bos_token_id": null,
|
| 78 |
+
"dtype": "bfloat16",
|
| 79 |
+
"eos_token_id": null,
|
| 80 |
+
"hidden_act": "silu",
|
| 81 |
+
"hidden_size": 3584,
|
| 82 |
+
"initializer_range": 0.02,
|
| 83 |
+
"intermediate_size": 18944,
|
| 84 |
+
"layer_types": [
|
| 85 |
+
"full_attention",
|
| 86 |
+
"full_attention",
|
| 87 |
+
"full_attention",
|
| 88 |
+
"full_attention",
|
| 89 |
+
"full_attention",
|
| 90 |
+
"full_attention",
|
| 91 |
+
"full_attention",
|
| 92 |
+
"full_attention",
|
| 93 |
+
"full_attention",
|
| 94 |
+
"full_attention",
|
| 95 |
+
"full_attention",
|
| 96 |
+
"full_attention",
|
| 97 |
+
"full_attention",
|
| 98 |
+
"full_attention",
|
| 99 |
+
"full_attention",
|
| 100 |
+
"full_attention",
|
| 101 |
+
"full_attention",
|
| 102 |
+
"full_attention",
|
| 103 |
+
"full_attention",
|
| 104 |
+
"full_attention",
|
| 105 |
+
"full_attention",
|
| 106 |
+
"full_attention",
|
| 107 |
+
"full_attention",
|
| 108 |
+
"full_attention",
|
| 109 |
+
"full_attention",
|
| 110 |
+
"full_attention",
|
| 111 |
+
"full_attention",
|
| 112 |
+
"full_attention"
|
| 113 |
+
],
|
| 114 |
+
"max_position_embeddings": 131072,
|
| 115 |
+
"max_window_layers": 28,
|
| 116 |
+
"model_type": "qwen2",
|
| 117 |
+
"num_attention_heads": 28,
|
| 118 |
+
"num_hidden_layers": 28,
|
| 119 |
+
"num_key_value_heads": 4,
|
| 120 |
+
"pad_token_id": null,
|
| 121 |
+
"rms_norm_eps": 1e-06,
|
| 122 |
+
"rope_parameters": {
|
| 123 |
+
"rope_theta": 1000000.0,
|
| 124 |
+
"rope_type": "default"
|
| 125 |
+
},
|
| 126 |
+
"sliding_window": null,
|
| 127 |
+
"tie_word_embeddings": false,
|
| 128 |
+
"use_cache": true,
|
| 129 |
+
"use_mrope": false,
|
| 130 |
+
"use_sliding_window": false,
|
| 131 |
+
"vocab_size": 152064
|
| 132 |
+
},
|
| 133 |
+
"transformers_version": "5.3.0.dev0"
|
| 134 |
+
}
|
figures/DER.jpg
ADDED
|
figures/VibeVoice_ASR_archi.png
ADDED
|
Git LFS Details
|
figures/cpWER.jpg
ADDED
|
figures/language_distribution_horizontal.png
ADDED
|
Git LFS Details
|
figures/tcpWER.jpg
ADDED
|
generation_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"do_sample": false,
|
| 4 |
+
"eos_token_id": 151643,
|
| 5 |
+
"max_length": 32768,
|
| 6 |
+
"max_new_tokens": 32768,
|
| 7 |
+
"output_attentions": false,
|
| 8 |
+
"output_hidden_states": false,
|
| 9 |
+
"pad_token_id": 151655,
|
| 10 |
+
"transformers_version": "5.3.0.dev0",
|
| 11 |
+
"use_cache": true
|
| 12 |
+
}
|
model-00001-of-00008.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:580189757a1c737ecc6fad16e633b922bc73220268356e31362c24365f081095
|
| 3 |
+
size 2488346304
|
model-00002-of-00008.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:65e419d28fc8c87d6938ec611c7b425b38e391c80a2df59c2b4935f6fd65c7c2
|
| 3 |
+
size 2389316008
|
model-00003-of-00008.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:30f574c764550d26c654cc3f5dd000b90b0c3305db955267c577001c1013fdf7
|
| 3 |
+
size 2466376400
|
model-00004-of-00008.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8fa3d9350f0a8cc97524617ff582e451159bfdba2ef0b2e372a627518ef681b1
|
| 3 |
+
size 2466376432
|
model-00005-of-00008.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5015ecacc86d897d20d5e42ba52ef83355d42f587d3b158ff897ec94847bd083
|
| 3 |
+
size 2499431160
|
model-00006-of-00008.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7463d75607185c118925cea6a60a5698aaa6f428e626bcee774572bdc28ef55d
|
| 3 |
+
size 1831435704
|
model-00007-of-00008.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b52aa2fab9640bcac5201a278f540b018c439ed4e99516d9ff30ea1ae399525e
|
| 3 |
+
size 2482226384
|
model-00008-of-00008.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a66ced85b619507e1970dc51be45786ff55803ccc63658765e412ea23d9a8ace
|
| 3 |
+
size 37262376
|
model.safetensors.index.json
ADDED
|
@@ -0,0 +1,909 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"metadata": {
|
| 3 |
+
"total_parameters": 8330325888,
|
| 4 |
+
"total_size": 16660651776
|
| 5 |
+
},
|
| 6 |
+
"weight_map": {
|
| 7 |
+
"acoustic_tokenizer_encoder.conv_layers.0.conv.conv.bias": "model-00007-of-00008.safetensors",
|
| 8 |
+
"acoustic_tokenizer_encoder.conv_layers.0.conv.conv.weight": "model-00007-of-00008.safetensors",
|
| 9 |
+
"acoustic_tokenizer_encoder.conv_layers.0.stage.0.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 10 |
+
"acoustic_tokenizer_encoder.conv_layers.0.stage.0.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 11 |
+
"acoustic_tokenizer_encoder.conv_layers.0.stage.0.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 12 |
+
"acoustic_tokenizer_encoder.conv_layers.0.stage.0.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 13 |
+
"acoustic_tokenizer_encoder.conv_layers.0.stage.0.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 14 |
+
"acoustic_tokenizer_encoder.conv_layers.0.stage.0.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 15 |
+
"acoustic_tokenizer_encoder.conv_layers.0.stage.0.gamma": "model-00007-of-00008.safetensors",
|
| 16 |
+
"acoustic_tokenizer_encoder.conv_layers.0.stage.0.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 17 |
+
"acoustic_tokenizer_encoder.conv_layers.0.stage.0.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 18 |
+
"acoustic_tokenizer_encoder.conv_layers.0.stage.0.norm.weight": "model-00007-of-00008.safetensors",
|
| 19 |
+
"acoustic_tokenizer_encoder.conv_layers.0.stage.1.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 20 |
+
"acoustic_tokenizer_encoder.conv_layers.0.stage.1.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 21 |
+
"acoustic_tokenizer_encoder.conv_layers.0.stage.1.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 22 |
+
"acoustic_tokenizer_encoder.conv_layers.0.stage.1.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 23 |
+
"acoustic_tokenizer_encoder.conv_layers.0.stage.1.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 24 |
+
"acoustic_tokenizer_encoder.conv_layers.0.stage.1.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 25 |
+
"acoustic_tokenizer_encoder.conv_layers.0.stage.1.gamma": "model-00007-of-00008.safetensors",
|
| 26 |
+
"acoustic_tokenizer_encoder.conv_layers.0.stage.1.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 27 |
+
"acoustic_tokenizer_encoder.conv_layers.0.stage.1.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 28 |
+
"acoustic_tokenizer_encoder.conv_layers.0.stage.1.norm.weight": "model-00007-of-00008.safetensors",
|
| 29 |
+
"acoustic_tokenizer_encoder.conv_layers.0.stage.2.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 30 |
+
"acoustic_tokenizer_encoder.conv_layers.0.stage.2.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 31 |
+
"acoustic_tokenizer_encoder.conv_layers.0.stage.2.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 32 |
+
"acoustic_tokenizer_encoder.conv_layers.0.stage.2.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 33 |
+
"acoustic_tokenizer_encoder.conv_layers.0.stage.2.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 34 |
+
"acoustic_tokenizer_encoder.conv_layers.0.stage.2.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 35 |
+
"acoustic_tokenizer_encoder.conv_layers.0.stage.2.gamma": "model-00007-of-00008.safetensors",
|
| 36 |
+
"acoustic_tokenizer_encoder.conv_layers.0.stage.2.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 37 |
+
"acoustic_tokenizer_encoder.conv_layers.0.stage.2.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 38 |
+
"acoustic_tokenizer_encoder.conv_layers.0.stage.2.norm.weight": "model-00007-of-00008.safetensors",
|
| 39 |
+
"acoustic_tokenizer_encoder.conv_layers.1.conv.conv.bias": "model-00007-of-00008.safetensors",
|
| 40 |
+
"acoustic_tokenizer_encoder.conv_layers.1.conv.conv.weight": "model-00007-of-00008.safetensors",
|
| 41 |
+
"acoustic_tokenizer_encoder.conv_layers.1.stage.0.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 42 |
+
"acoustic_tokenizer_encoder.conv_layers.1.stage.0.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 43 |
+
"acoustic_tokenizer_encoder.conv_layers.1.stage.0.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 44 |
+
"acoustic_tokenizer_encoder.conv_layers.1.stage.0.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 45 |
+
"acoustic_tokenizer_encoder.conv_layers.1.stage.0.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 46 |
+
"acoustic_tokenizer_encoder.conv_layers.1.stage.0.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 47 |
+
"acoustic_tokenizer_encoder.conv_layers.1.stage.0.gamma": "model-00007-of-00008.safetensors",
|
| 48 |
+
"acoustic_tokenizer_encoder.conv_layers.1.stage.0.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 49 |
+
"acoustic_tokenizer_encoder.conv_layers.1.stage.0.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 50 |
+
"acoustic_tokenizer_encoder.conv_layers.1.stage.0.norm.weight": "model-00007-of-00008.safetensors",
|
| 51 |
+
"acoustic_tokenizer_encoder.conv_layers.1.stage.1.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 52 |
+
"acoustic_tokenizer_encoder.conv_layers.1.stage.1.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 53 |
+
"acoustic_tokenizer_encoder.conv_layers.1.stage.1.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 54 |
+
"acoustic_tokenizer_encoder.conv_layers.1.stage.1.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 55 |
+
"acoustic_tokenizer_encoder.conv_layers.1.stage.1.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 56 |
+
"acoustic_tokenizer_encoder.conv_layers.1.stage.1.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 57 |
+
"acoustic_tokenizer_encoder.conv_layers.1.stage.1.gamma": "model-00007-of-00008.safetensors",
|
| 58 |
+
"acoustic_tokenizer_encoder.conv_layers.1.stage.1.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 59 |
+
"acoustic_tokenizer_encoder.conv_layers.1.stage.1.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 60 |
+
"acoustic_tokenizer_encoder.conv_layers.1.stage.1.norm.weight": "model-00007-of-00008.safetensors",
|
| 61 |
+
"acoustic_tokenizer_encoder.conv_layers.1.stage.2.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 62 |
+
"acoustic_tokenizer_encoder.conv_layers.1.stage.2.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 63 |
+
"acoustic_tokenizer_encoder.conv_layers.1.stage.2.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 64 |
+
"acoustic_tokenizer_encoder.conv_layers.1.stage.2.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 65 |
+
"acoustic_tokenizer_encoder.conv_layers.1.stage.2.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 66 |
+
"acoustic_tokenizer_encoder.conv_layers.1.stage.2.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 67 |
+
"acoustic_tokenizer_encoder.conv_layers.1.stage.2.gamma": "model-00007-of-00008.safetensors",
|
| 68 |
+
"acoustic_tokenizer_encoder.conv_layers.1.stage.2.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 69 |
+
"acoustic_tokenizer_encoder.conv_layers.1.stage.2.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 70 |
+
"acoustic_tokenizer_encoder.conv_layers.1.stage.2.norm.weight": "model-00007-of-00008.safetensors",
|
| 71 |
+
"acoustic_tokenizer_encoder.conv_layers.2.conv.conv.bias": "model-00007-of-00008.safetensors",
|
| 72 |
+
"acoustic_tokenizer_encoder.conv_layers.2.conv.conv.weight": "model-00007-of-00008.safetensors",
|
| 73 |
+
"acoustic_tokenizer_encoder.conv_layers.2.stage.0.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 74 |
+
"acoustic_tokenizer_encoder.conv_layers.2.stage.0.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 75 |
+
"acoustic_tokenizer_encoder.conv_layers.2.stage.0.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 76 |
+
"acoustic_tokenizer_encoder.conv_layers.2.stage.0.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 77 |
+
"acoustic_tokenizer_encoder.conv_layers.2.stage.0.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 78 |
+
"acoustic_tokenizer_encoder.conv_layers.2.stage.0.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 79 |
+
"acoustic_tokenizer_encoder.conv_layers.2.stage.0.gamma": "model-00007-of-00008.safetensors",
|
| 80 |
+
"acoustic_tokenizer_encoder.conv_layers.2.stage.0.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 81 |
+
"acoustic_tokenizer_encoder.conv_layers.2.stage.0.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 82 |
+
"acoustic_tokenizer_encoder.conv_layers.2.stage.0.norm.weight": "model-00007-of-00008.safetensors",
|
| 83 |
+
"acoustic_tokenizer_encoder.conv_layers.2.stage.1.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 84 |
+
"acoustic_tokenizer_encoder.conv_layers.2.stage.1.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 85 |
+
"acoustic_tokenizer_encoder.conv_layers.2.stage.1.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 86 |
+
"acoustic_tokenizer_encoder.conv_layers.2.stage.1.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 87 |
+
"acoustic_tokenizer_encoder.conv_layers.2.stage.1.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 88 |
+
"acoustic_tokenizer_encoder.conv_layers.2.stage.1.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 89 |
+
"acoustic_tokenizer_encoder.conv_layers.2.stage.1.gamma": "model-00007-of-00008.safetensors",
|
| 90 |
+
"acoustic_tokenizer_encoder.conv_layers.2.stage.1.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 91 |
+
"acoustic_tokenizer_encoder.conv_layers.2.stage.1.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 92 |
+
"acoustic_tokenizer_encoder.conv_layers.2.stage.1.norm.weight": "model-00007-of-00008.safetensors",
|
| 93 |
+
"acoustic_tokenizer_encoder.conv_layers.2.stage.2.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 94 |
+
"acoustic_tokenizer_encoder.conv_layers.2.stage.2.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 95 |
+
"acoustic_tokenizer_encoder.conv_layers.2.stage.2.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 96 |
+
"acoustic_tokenizer_encoder.conv_layers.2.stage.2.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 97 |
+
"acoustic_tokenizer_encoder.conv_layers.2.stage.2.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 98 |
+
"acoustic_tokenizer_encoder.conv_layers.2.stage.2.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 99 |
+
"acoustic_tokenizer_encoder.conv_layers.2.stage.2.gamma": "model-00007-of-00008.safetensors",
|
| 100 |
+
"acoustic_tokenizer_encoder.conv_layers.2.stage.2.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 101 |
+
"acoustic_tokenizer_encoder.conv_layers.2.stage.2.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 102 |
+
"acoustic_tokenizer_encoder.conv_layers.2.stage.2.norm.weight": "model-00007-of-00008.safetensors",
|
| 103 |
+
"acoustic_tokenizer_encoder.conv_layers.3.conv.conv.bias": "model-00007-of-00008.safetensors",
|
| 104 |
+
"acoustic_tokenizer_encoder.conv_layers.3.conv.conv.weight": "model-00007-of-00008.safetensors",
|
| 105 |
+
"acoustic_tokenizer_encoder.conv_layers.3.stage.0.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 106 |
+
"acoustic_tokenizer_encoder.conv_layers.3.stage.0.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 107 |
+
"acoustic_tokenizer_encoder.conv_layers.3.stage.0.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 108 |
+
"acoustic_tokenizer_encoder.conv_layers.3.stage.0.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 109 |
+
"acoustic_tokenizer_encoder.conv_layers.3.stage.0.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 110 |
+
"acoustic_tokenizer_encoder.conv_layers.3.stage.0.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 111 |
+
"acoustic_tokenizer_encoder.conv_layers.3.stage.0.gamma": "model-00007-of-00008.safetensors",
|
| 112 |
+
"acoustic_tokenizer_encoder.conv_layers.3.stage.0.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 113 |
+
"acoustic_tokenizer_encoder.conv_layers.3.stage.0.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 114 |
+
"acoustic_tokenizer_encoder.conv_layers.3.stage.0.norm.weight": "model-00007-of-00008.safetensors",
|
| 115 |
+
"acoustic_tokenizer_encoder.conv_layers.3.stage.1.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 116 |
+
"acoustic_tokenizer_encoder.conv_layers.3.stage.1.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 117 |
+
"acoustic_tokenizer_encoder.conv_layers.3.stage.1.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 118 |
+
"acoustic_tokenizer_encoder.conv_layers.3.stage.1.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 119 |
+
"acoustic_tokenizer_encoder.conv_layers.3.stage.1.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 120 |
+
"acoustic_tokenizer_encoder.conv_layers.3.stage.1.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 121 |
+
"acoustic_tokenizer_encoder.conv_layers.3.stage.1.gamma": "model-00007-of-00008.safetensors",
|
| 122 |
+
"acoustic_tokenizer_encoder.conv_layers.3.stage.1.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 123 |
+
"acoustic_tokenizer_encoder.conv_layers.3.stage.1.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 124 |
+
"acoustic_tokenizer_encoder.conv_layers.3.stage.1.norm.weight": "model-00007-of-00008.safetensors",
|
| 125 |
+
"acoustic_tokenizer_encoder.conv_layers.3.stage.2.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 126 |
+
"acoustic_tokenizer_encoder.conv_layers.3.stage.2.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 127 |
+
"acoustic_tokenizer_encoder.conv_layers.3.stage.2.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 128 |
+
"acoustic_tokenizer_encoder.conv_layers.3.stage.2.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 129 |
+
"acoustic_tokenizer_encoder.conv_layers.3.stage.2.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 130 |
+
"acoustic_tokenizer_encoder.conv_layers.3.stage.2.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 131 |
+
"acoustic_tokenizer_encoder.conv_layers.3.stage.2.gamma": "model-00007-of-00008.safetensors",
|
| 132 |
+
"acoustic_tokenizer_encoder.conv_layers.3.stage.2.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 133 |
+
"acoustic_tokenizer_encoder.conv_layers.3.stage.2.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 134 |
+
"acoustic_tokenizer_encoder.conv_layers.3.stage.2.norm.weight": "model-00007-of-00008.safetensors",
|
| 135 |
+
"acoustic_tokenizer_encoder.conv_layers.4.conv.conv.bias": "model-00007-of-00008.safetensors",
|
| 136 |
+
"acoustic_tokenizer_encoder.conv_layers.4.conv.conv.weight": "model-00007-of-00008.safetensors",
|
| 137 |
+
"acoustic_tokenizer_encoder.conv_layers.4.stage.0.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 138 |
+
"acoustic_tokenizer_encoder.conv_layers.4.stage.0.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 139 |
+
"acoustic_tokenizer_encoder.conv_layers.4.stage.0.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 140 |
+
"acoustic_tokenizer_encoder.conv_layers.4.stage.0.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 141 |
+
"acoustic_tokenizer_encoder.conv_layers.4.stage.0.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 142 |
+
"acoustic_tokenizer_encoder.conv_layers.4.stage.0.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 143 |
+
"acoustic_tokenizer_encoder.conv_layers.4.stage.0.gamma": "model-00007-of-00008.safetensors",
|
| 144 |
+
"acoustic_tokenizer_encoder.conv_layers.4.stage.0.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 145 |
+
"acoustic_tokenizer_encoder.conv_layers.4.stage.0.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 146 |
+
"acoustic_tokenizer_encoder.conv_layers.4.stage.0.norm.weight": "model-00007-of-00008.safetensors",
|
| 147 |
+
"acoustic_tokenizer_encoder.conv_layers.4.stage.1.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 148 |
+
"acoustic_tokenizer_encoder.conv_layers.4.stage.1.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 149 |
+
"acoustic_tokenizer_encoder.conv_layers.4.stage.1.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 150 |
+
"acoustic_tokenizer_encoder.conv_layers.4.stage.1.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 151 |
+
"acoustic_tokenizer_encoder.conv_layers.4.stage.1.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 152 |
+
"acoustic_tokenizer_encoder.conv_layers.4.stage.1.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 153 |
+
"acoustic_tokenizer_encoder.conv_layers.4.stage.1.gamma": "model-00007-of-00008.safetensors",
|
| 154 |
+
"acoustic_tokenizer_encoder.conv_layers.4.stage.1.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 155 |
+
"acoustic_tokenizer_encoder.conv_layers.4.stage.1.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 156 |
+
"acoustic_tokenizer_encoder.conv_layers.4.stage.1.norm.weight": "model-00007-of-00008.safetensors",
|
| 157 |
+
"acoustic_tokenizer_encoder.conv_layers.4.stage.2.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 158 |
+
"acoustic_tokenizer_encoder.conv_layers.4.stage.2.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 159 |
+
"acoustic_tokenizer_encoder.conv_layers.4.stage.2.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 160 |
+
"acoustic_tokenizer_encoder.conv_layers.4.stage.2.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 161 |
+
"acoustic_tokenizer_encoder.conv_layers.4.stage.2.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 162 |
+
"acoustic_tokenizer_encoder.conv_layers.4.stage.2.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 163 |
+
"acoustic_tokenizer_encoder.conv_layers.4.stage.2.gamma": "model-00007-of-00008.safetensors",
|
| 164 |
+
"acoustic_tokenizer_encoder.conv_layers.4.stage.2.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 165 |
+
"acoustic_tokenizer_encoder.conv_layers.4.stage.2.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 166 |
+
"acoustic_tokenizer_encoder.conv_layers.4.stage.2.norm.weight": "model-00007-of-00008.safetensors",
|
| 167 |
+
"acoustic_tokenizer_encoder.conv_layers.5.conv.conv.bias": "model-00007-of-00008.safetensors",
|
| 168 |
+
"acoustic_tokenizer_encoder.conv_layers.5.conv.conv.weight": "model-00007-of-00008.safetensors",
|
| 169 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.0.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 170 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.0.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 171 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.0.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 172 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.0.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 173 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.0.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 174 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.0.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 175 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.0.gamma": "model-00007-of-00008.safetensors",
|
| 176 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.0.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 177 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.0.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 178 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.0.norm.weight": "model-00007-of-00008.safetensors",
|
| 179 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.1.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 180 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.1.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 181 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.1.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 182 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.1.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 183 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.1.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 184 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.1.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 185 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.1.gamma": "model-00007-of-00008.safetensors",
|
| 186 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.1.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 187 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.1.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 188 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.1.norm.weight": "model-00007-of-00008.safetensors",
|
| 189 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.2.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 190 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.2.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 191 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.2.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 192 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.2.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 193 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.2.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 194 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.2.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 195 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.2.gamma": "model-00007-of-00008.safetensors",
|
| 196 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.2.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 197 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.2.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 198 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.2.norm.weight": "model-00007-of-00008.safetensors",
|
| 199 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.3.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 200 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.3.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 201 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.3.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 202 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.3.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 203 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.3.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 204 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.3.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 205 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.3.gamma": "model-00007-of-00008.safetensors",
|
| 206 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.3.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 207 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.3.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 208 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.3.norm.weight": "model-00007-of-00008.safetensors",
|
| 209 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.4.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 210 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.4.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 211 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.4.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 212 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.4.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 213 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.4.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 214 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.4.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 215 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.4.gamma": "model-00007-of-00008.safetensors",
|
| 216 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.4.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 217 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.4.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 218 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.4.norm.weight": "model-00007-of-00008.safetensors",
|
| 219 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.5.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 220 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.5.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 221 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.5.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 222 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.5.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 223 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.5.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 224 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.5.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 225 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.5.gamma": "model-00007-of-00008.safetensors",
|
| 226 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.5.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 227 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.5.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 228 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.5.norm.weight": "model-00007-of-00008.safetensors",
|
| 229 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.6.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 230 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.6.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 231 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.6.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 232 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.6.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 233 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.6.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 234 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.6.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 235 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.6.gamma": "model-00007-of-00008.safetensors",
|
| 236 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.6.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 237 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.6.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 238 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.6.norm.weight": "model-00007-of-00008.safetensors",
|
| 239 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.7.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 240 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.7.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 241 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.7.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 242 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.7.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 243 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.7.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 244 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.7.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 245 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.7.gamma": "model-00007-of-00008.safetensors",
|
| 246 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.7.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 247 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.7.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 248 |
+
"acoustic_tokenizer_encoder.conv_layers.5.stage.7.norm.weight": "model-00007-of-00008.safetensors",
|
| 249 |
+
"acoustic_tokenizer_encoder.head.conv.bias": "model-00007-of-00008.safetensors",
|
| 250 |
+
"acoustic_tokenizer_encoder.head.conv.weight": "model-00007-of-00008.safetensors",
|
| 251 |
+
"acoustic_tokenizer_encoder.stem.conv.conv.bias": "model-00007-of-00008.safetensors",
|
| 252 |
+
"acoustic_tokenizer_encoder.stem.conv.conv.weight": "model-00007-of-00008.safetensors",
|
| 253 |
+
"acoustic_tokenizer_encoder.stem.stage.0.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 254 |
+
"acoustic_tokenizer_encoder.stem.stage.0.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 255 |
+
"acoustic_tokenizer_encoder.stem.stage.0.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 256 |
+
"acoustic_tokenizer_encoder.stem.stage.0.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 257 |
+
"acoustic_tokenizer_encoder.stem.stage.0.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 258 |
+
"acoustic_tokenizer_encoder.stem.stage.0.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 259 |
+
"acoustic_tokenizer_encoder.stem.stage.0.gamma": "model-00007-of-00008.safetensors",
|
| 260 |
+
"acoustic_tokenizer_encoder.stem.stage.0.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 261 |
+
"acoustic_tokenizer_encoder.stem.stage.0.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 262 |
+
"acoustic_tokenizer_encoder.stem.stage.0.norm.weight": "model-00007-of-00008.safetensors",
|
| 263 |
+
"acoustic_tokenizer_encoder.stem.stage.1.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 264 |
+
"acoustic_tokenizer_encoder.stem.stage.1.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 265 |
+
"acoustic_tokenizer_encoder.stem.stage.1.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 266 |
+
"acoustic_tokenizer_encoder.stem.stage.1.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 267 |
+
"acoustic_tokenizer_encoder.stem.stage.1.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 268 |
+
"acoustic_tokenizer_encoder.stem.stage.1.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 269 |
+
"acoustic_tokenizer_encoder.stem.stage.1.gamma": "model-00007-of-00008.safetensors",
|
| 270 |
+
"acoustic_tokenizer_encoder.stem.stage.1.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 271 |
+
"acoustic_tokenizer_encoder.stem.stage.1.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 272 |
+
"acoustic_tokenizer_encoder.stem.stage.1.norm.weight": "model-00007-of-00008.safetensors",
|
| 273 |
+
"acoustic_tokenizer_encoder.stem.stage.2.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 274 |
+
"acoustic_tokenizer_encoder.stem.stage.2.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 275 |
+
"acoustic_tokenizer_encoder.stem.stage.2.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 276 |
+
"acoustic_tokenizer_encoder.stem.stage.2.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 277 |
+
"acoustic_tokenizer_encoder.stem.stage.2.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 278 |
+
"acoustic_tokenizer_encoder.stem.stage.2.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 279 |
+
"acoustic_tokenizer_encoder.stem.stage.2.gamma": "model-00007-of-00008.safetensors",
|
| 280 |
+
"acoustic_tokenizer_encoder.stem.stage.2.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 281 |
+
"acoustic_tokenizer_encoder.stem.stage.2.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 282 |
+
"acoustic_tokenizer_encoder.stem.stage.2.norm.weight": "model-00007-of-00008.safetensors",
|
| 283 |
+
"language_model.lm_head.weight": "model-00007-of-00008.safetensors",
|
| 284 |
+
"language_model.model.embed_tokens.weight": "model-00001-of-00008.safetensors",
|
| 285 |
+
"language_model.model.layers.0.input_layernorm.weight": "model-00001-of-00008.safetensors",
|
| 286 |
+
"language_model.model.layers.0.mlp.down_proj.weight": "model-00001-of-00008.safetensors",
|
| 287 |
+
"language_model.model.layers.0.mlp.gate_proj.weight": "model-00001-of-00008.safetensors",
|
| 288 |
+
"language_model.model.layers.0.mlp.up_proj.weight": "model-00001-of-00008.safetensors",
|
| 289 |
+
"language_model.model.layers.0.post_attention_layernorm.weight": "model-00001-of-00008.safetensors",
|
| 290 |
+
"language_model.model.layers.0.self_attn.k_proj.bias": "model-00001-of-00008.safetensors",
|
| 291 |
+
"language_model.model.layers.0.self_attn.k_proj.weight": "model-00001-of-00008.safetensors",
|
| 292 |
+
"language_model.model.layers.0.self_attn.o_proj.weight": "model-00001-of-00008.safetensors",
|
| 293 |
+
"language_model.model.layers.0.self_attn.q_proj.bias": "model-00001-of-00008.safetensors",
|
| 294 |
+
"language_model.model.layers.0.self_attn.q_proj.weight": "model-00001-of-00008.safetensors",
|
| 295 |
+
"language_model.model.layers.0.self_attn.v_proj.bias": "model-00001-of-00008.safetensors",
|
| 296 |
+
"language_model.model.layers.0.self_attn.v_proj.weight": "model-00001-of-00008.safetensors",
|
| 297 |
+
"language_model.model.layers.1.input_layernorm.weight": "model-00001-of-00008.safetensors",
|
| 298 |
+
"language_model.model.layers.1.mlp.down_proj.weight": "model-00001-of-00008.safetensors",
|
| 299 |
+
"language_model.model.layers.1.mlp.gate_proj.weight": "model-00001-of-00008.safetensors",
|
| 300 |
+
"language_model.model.layers.1.mlp.up_proj.weight": "model-00001-of-00008.safetensors",
|
| 301 |
+
"language_model.model.layers.1.post_attention_layernorm.weight": "model-00001-of-00008.safetensors",
|
| 302 |
+
"language_model.model.layers.1.self_attn.k_proj.bias": "model-00001-of-00008.safetensors",
|
| 303 |
+
"language_model.model.layers.1.self_attn.k_proj.weight": "model-00001-of-00008.safetensors",
|
| 304 |
+
"language_model.model.layers.1.self_attn.o_proj.weight": "model-00001-of-00008.safetensors",
|
| 305 |
+
"language_model.model.layers.1.self_attn.q_proj.bias": "model-00001-of-00008.safetensors",
|
| 306 |
+
"language_model.model.layers.1.self_attn.q_proj.weight": "model-00001-of-00008.safetensors",
|
| 307 |
+
"language_model.model.layers.1.self_attn.v_proj.bias": "model-00001-of-00008.safetensors",
|
| 308 |
+
"language_model.model.layers.1.self_attn.v_proj.weight": "model-00001-of-00008.safetensors",
|
| 309 |
+
"language_model.model.layers.10.input_layernorm.weight": "model-00003-of-00008.safetensors",
|
| 310 |
+
"language_model.model.layers.10.mlp.down_proj.weight": "model-00003-of-00008.safetensors",
|
| 311 |
+
"language_model.model.layers.10.mlp.gate_proj.weight": "model-00003-of-00008.safetensors",
|
| 312 |
+
"language_model.model.layers.10.mlp.up_proj.weight": "model-00003-of-00008.safetensors",
|
| 313 |
+
"language_model.model.layers.10.post_attention_layernorm.weight": "model-00003-of-00008.safetensors",
|
| 314 |
+
"language_model.model.layers.10.self_attn.k_proj.bias": "model-00003-of-00008.safetensors",
|
| 315 |
+
"language_model.model.layers.10.self_attn.k_proj.weight": "model-00003-of-00008.safetensors",
|
| 316 |
+
"language_model.model.layers.10.self_attn.o_proj.weight": "model-00003-of-00008.safetensors",
|
| 317 |
+
"language_model.model.layers.10.self_attn.q_proj.bias": "model-00003-of-00008.safetensors",
|
| 318 |
+
"language_model.model.layers.10.self_attn.q_proj.weight": "model-00003-of-00008.safetensors",
|
| 319 |
+
"language_model.model.layers.10.self_attn.v_proj.bias": "model-00003-of-00008.safetensors",
|
| 320 |
+
"language_model.model.layers.10.self_attn.v_proj.weight": "model-00003-of-00008.safetensors",
|
| 321 |
+
"language_model.model.layers.11.input_layernorm.weight": "model-00003-of-00008.safetensors",
|
| 322 |
+
"language_model.model.layers.11.mlp.down_proj.weight": "model-00003-of-00008.safetensors",
|
| 323 |
+
"language_model.model.layers.11.mlp.gate_proj.weight": "model-00003-of-00008.safetensors",
|
| 324 |
+
"language_model.model.layers.11.mlp.up_proj.weight": "model-00003-of-00008.safetensors",
|
| 325 |
+
"language_model.model.layers.11.post_attention_layernorm.weight": "model-00003-of-00008.safetensors",
|
| 326 |
+
"language_model.model.layers.11.self_attn.k_proj.bias": "model-00003-of-00008.safetensors",
|
| 327 |
+
"language_model.model.layers.11.self_attn.k_proj.weight": "model-00003-of-00008.safetensors",
|
| 328 |
+
"language_model.model.layers.11.self_attn.o_proj.weight": "model-00003-of-00008.safetensors",
|
| 329 |
+
"language_model.model.layers.11.self_attn.q_proj.bias": "model-00003-of-00008.safetensors",
|
| 330 |
+
"language_model.model.layers.11.self_attn.q_proj.weight": "model-00003-of-00008.safetensors",
|
| 331 |
+
"language_model.model.layers.11.self_attn.v_proj.bias": "model-00003-of-00008.safetensors",
|
| 332 |
+
"language_model.model.layers.11.self_attn.v_proj.weight": "model-00003-of-00008.safetensors",
|
| 333 |
+
"language_model.model.layers.12.input_layernorm.weight": "model-00003-of-00008.safetensors",
|
| 334 |
+
"language_model.model.layers.12.mlp.down_proj.weight": "model-00003-of-00008.safetensors",
|
| 335 |
+
"language_model.model.layers.12.mlp.gate_proj.weight": "model-00003-of-00008.safetensors",
|
| 336 |
+
"language_model.model.layers.12.mlp.up_proj.weight": "model-00003-of-00008.safetensors",
|
| 337 |
+
"language_model.model.layers.12.post_attention_layernorm.weight": "model-00003-of-00008.safetensors",
|
| 338 |
+
"language_model.model.layers.12.self_attn.k_proj.bias": "model-00003-of-00008.safetensors",
|
| 339 |
+
"language_model.model.layers.12.self_attn.k_proj.weight": "model-00003-of-00008.safetensors",
|
| 340 |
+
"language_model.model.layers.12.self_attn.o_proj.weight": "model-00003-of-00008.safetensors",
|
| 341 |
+
"language_model.model.layers.12.self_attn.q_proj.bias": "model-00003-of-00008.safetensors",
|
| 342 |
+
"language_model.model.layers.12.self_attn.q_proj.weight": "model-00003-of-00008.safetensors",
|
| 343 |
+
"language_model.model.layers.12.self_attn.v_proj.bias": "model-00003-of-00008.safetensors",
|
| 344 |
+
"language_model.model.layers.12.self_attn.v_proj.weight": "model-00003-of-00008.safetensors",
|
| 345 |
+
"language_model.model.layers.13.input_layernorm.weight": "model-00004-of-00008.safetensors",
|
| 346 |
+
"language_model.model.layers.13.mlp.down_proj.weight": "model-00004-of-00008.safetensors",
|
| 347 |
+
"language_model.model.layers.13.mlp.gate_proj.weight": "model-00003-of-00008.safetensors",
|
| 348 |
+
"language_model.model.layers.13.mlp.up_proj.weight": "model-00004-of-00008.safetensors",
|
| 349 |
+
"language_model.model.layers.13.post_attention_layernorm.weight": "model-00004-of-00008.safetensors",
|
| 350 |
+
"language_model.model.layers.13.self_attn.k_proj.bias": "model-00003-of-00008.safetensors",
|
| 351 |
+
"language_model.model.layers.13.self_attn.k_proj.weight": "model-00003-of-00008.safetensors",
|
| 352 |
+
"language_model.model.layers.13.self_attn.o_proj.weight": "model-00003-of-00008.safetensors",
|
| 353 |
+
"language_model.model.layers.13.self_attn.q_proj.bias": "model-00003-of-00008.safetensors",
|
| 354 |
+
"language_model.model.layers.13.self_attn.q_proj.weight": "model-00003-of-00008.safetensors",
|
| 355 |
+
"language_model.model.layers.13.self_attn.v_proj.bias": "model-00003-of-00008.safetensors",
|
| 356 |
+
"language_model.model.layers.13.self_attn.v_proj.weight": "model-00003-of-00008.safetensors",
|
| 357 |
+
"language_model.model.layers.14.input_layernorm.weight": "model-00004-of-00008.safetensors",
|
| 358 |
+
"language_model.model.layers.14.mlp.down_proj.weight": "model-00004-of-00008.safetensors",
|
| 359 |
+
"language_model.model.layers.14.mlp.gate_proj.weight": "model-00004-of-00008.safetensors",
|
| 360 |
+
"language_model.model.layers.14.mlp.up_proj.weight": "model-00004-of-00008.safetensors",
|
| 361 |
+
"language_model.model.layers.14.post_attention_layernorm.weight": "model-00004-of-00008.safetensors",
|
| 362 |
+
"language_model.model.layers.14.self_attn.k_proj.bias": "model-00004-of-00008.safetensors",
|
| 363 |
+
"language_model.model.layers.14.self_attn.k_proj.weight": "model-00004-of-00008.safetensors",
|
| 364 |
+
"language_model.model.layers.14.self_attn.o_proj.weight": "model-00004-of-00008.safetensors",
|
| 365 |
+
"language_model.model.layers.14.self_attn.q_proj.bias": "model-00004-of-00008.safetensors",
|
| 366 |
+
"language_model.model.layers.14.self_attn.q_proj.weight": "model-00004-of-00008.safetensors",
|
| 367 |
+
"language_model.model.layers.14.self_attn.v_proj.bias": "model-00004-of-00008.safetensors",
|
| 368 |
+
"language_model.model.layers.14.self_attn.v_proj.weight": "model-00004-of-00008.safetensors",
|
| 369 |
+
"language_model.model.layers.15.input_layernorm.weight": "model-00004-of-00008.safetensors",
|
| 370 |
+
"language_model.model.layers.15.mlp.down_proj.weight": "model-00004-of-00008.safetensors",
|
| 371 |
+
"language_model.model.layers.15.mlp.gate_proj.weight": "model-00004-of-00008.safetensors",
|
| 372 |
+
"language_model.model.layers.15.mlp.up_proj.weight": "model-00004-of-00008.safetensors",
|
| 373 |
+
"language_model.model.layers.15.post_attention_layernorm.weight": "model-00004-of-00008.safetensors",
|
| 374 |
+
"language_model.model.layers.15.self_attn.k_proj.bias": "model-00004-of-00008.safetensors",
|
| 375 |
+
"language_model.model.layers.15.self_attn.k_proj.weight": "model-00004-of-00008.safetensors",
|
| 376 |
+
"language_model.model.layers.15.self_attn.o_proj.weight": "model-00004-of-00008.safetensors",
|
| 377 |
+
"language_model.model.layers.15.self_attn.q_proj.bias": "model-00004-of-00008.safetensors",
|
| 378 |
+
"language_model.model.layers.15.self_attn.q_proj.weight": "model-00004-of-00008.safetensors",
|
| 379 |
+
"language_model.model.layers.15.self_attn.v_proj.bias": "model-00004-of-00008.safetensors",
|
| 380 |
+
"language_model.model.layers.15.self_attn.v_proj.weight": "model-00004-of-00008.safetensors",
|
| 381 |
+
"language_model.model.layers.16.input_layernorm.weight": "model-00004-of-00008.safetensors",
|
| 382 |
+
"language_model.model.layers.16.mlp.down_proj.weight": "model-00004-of-00008.safetensors",
|
| 383 |
+
"language_model.model.layers.16.mlp.gate_proj.weight": "model-00004-of-00008.safetensors",
|
| 384 |
+
"language_model.model.layers.16.mlp.up_proj.weight": "model-00004-of-00008.safetensors",
|
| 385 |
+
"language_model.model.layers.16.post_attention_layernorm.weight": "model-00004-of-00008.safetensors",
|
| 386 |
+
"language_model.model.layers.16.self_attn.k_proj.bias": "model-00004-of-00008.safetensors",
|
| 387 |
+
"language_model.model.layers.16.self_attn.k_proj.weight": "model-00004-of-00008.safetensors",
|
| 388 |
+
"language_model.model.layers.16.self_attn.o_proj.weight": "model-00004-of-00008.safetensors",
|
| 389 |
+
"language_model.model.layers.16.self_attn.q_proj.bias": "model-00004-of-00008.safetensors",
|
| 390 |
+
"language_model.model.layers.16.self_attn.q_proj.weight": "model-00004-of-00008.safetensors",
|
| 391 |
+
"language_model.model.layers.16.self_attn.v_proj.bias": "model-00004-of-00008.safetensors",
|
| 392 |
+
"language_model.model.layers.16.self_attn.v_proj.weight": "model-00004-of-00008.safetensors",
|
| 393 |
+
"language_model.model.layers.17.input_layernorm.weight": "model-00004-of-00008.safetensors",
|
| 394 |
+
"language_model.model.layers.17.mlp.down_proj.weight": "model-00004-of-00008.safetensors",
|
| 395 |
+
"language_model.model.layers.17.mlp.gate_proj.weight": "model-00004-of-00008.safetensors",
|
| 396 |
+
"language_model.model.layers.17.mlp.up_proj.weight": "model-00004-of-00008.safetensors",
|
| 397 |
+
"language_model.model.layers.17.post_attention_layernorm.weight": "model-00004-of-00008.safetensors",
|
| 398 |
+
"language_model.model.layers.17.self_attn.k_proj.bias": "model-00004-of-00008.safetensors",
|
| 399 |
+
"language_model.model.layers.17.self_attn.k_proj.weight": "model-00004-of-00008.safetensors",
|
| 400 |
+
"language_model.model.layers.17.self_attn.o_proj.weight": "model-00004-of-00008.safetensors",
|
| 401 |
+
"language_model.model.layers.17.self_attn.q_proj.bias": "model-00004-of-00008.safetensors",
|
| 402 |
+
"language_model.model.layers.17.self_attn.q_proj.weight": "model-00004-of-00008.safetensors",
|
| 403 |
+
"language_model.model.layers.17.self_attn.v_proj.bias": "model-00004-of-00008.safetensors",
|
| 404 |
+
"language_model.model.layers.17.self_attn.v_proj.weight": "model-00004-of-00008.safetensors",
|
| 405 |
+
"language_model.model.layers.18.input_layernorm.weight": "model-00005-of-00008.safetensors",
|
| 406 |
+
"language_model.model.layers.18.mlp.down_proj.weight": "model-00005-of-00008.safetensors",
|
| 407 |
+
"language_model.model.layers.18.mlp.gate_proj.weight": "model-00004-of-00008.safetensors",
|
| 408 |
+
"language_model.model.layers.18.mlp.up_proj.weight": "model-00004-of-00008.safetensors",
|
| 409 |
+
"language_model.model.layers.18.post_attention_layernorm.weight": "model-00005-of-00008.safetensors",
|
| 410 |
+
"language_model.model.layers.18.self_attn.k_proj.bias": "model-00004-of-00008.safetensors",
|
| 411 |
+
"language_model.model.layers.18.self_attn.k_proj.weight": "model-00004-of-00008.safetensors",
|
| 412 |
+
"language_model.model.layers.18.self_attn.o_proj.weight": "model-00004-of-00008.safetensors",
|
| 413 |
+
"language_model.model.layers.18.self_attn.q_proj.bias": "model-00004-of-00008.safetensors",
|
| 414 |
+
"language_model.model.layers.18.self_attn.q_proj.weight": "model-00004-of-00008.safetensors",
|
| 415 |
+
"language_model.model.layers.18.self_attn.v_proj.bias": "model-00004-of-00008.safetensors",
|
| 416 |
+
"language_model.model.layers.18.self_attn.v_proj.weight": "model-00004-of-00008.safetensors",
|
| 417 |
+
"language_model.model.layers.19.input_layernorm.weight": "model-00005-of-00008.safetensors",
|
| 418 |
+
"language_model.model.layers.19.mlp.down_proj.weight": "model-00005-of-00008.safetensors",
|
| 419 |
+
"language_model.model.layers.19.mlp.gate_proj.weight": "model-00005-of-00008.safetensors",
|
| 420 |
+
"language_model.model.layers.19.mlp.up_proj.weight": "model-00005-of-00008.safetensors",
|
| 421 |
+
"language_model.model.layers.19.post_attention_layernorm.weight": "model-00005-of-00008.safetensors",
|
| 422 |
+
"language_model.model.layers.19.self_attn.k_proj.bias": "model-00005-of-00008.safetensors",
|
| 423 |
+
"language_model.model.layers.19.self_attn.k_proj.weight": "model-00005-of-00008.safetensors",
|
| 424 |
+
"language_model.model.layers.19.self_attn.o_proj.weight": "model-00005-of-00008.safetensors",
|
| 425 |
+
"language_model.model.layers.19.self_attn.q_proj.bias": "model-00005-of-00008.safetensors",
|
| 426 |
+
"language_model.model.layers.19.self_attn.q_proj.weight": "model-00005-of-00008.safetensors",
|
| 427 |
+
"language_model.model.layers.19.self_attn.v_proj.bias": "model-00005-of-00008.safetensors",
|
| 428 |
+
"language_model.model.layers.19.self_attn.v_proj.weight": "model-00005-of-00008.safetensors",
|
| 429 |
+
"language_model.model.layers.2.input_layernorm.weight": "model-00001-of-00008.safetensors",
|
| 430 |
+
"language_model.model.layers.2.mlp.down_proj.weight": "model-00001-of-00008.safetensors",
|
| 431 |
+
"language_model.model.layers.2.mlp.gate_proj.weight": "model-00001-of-00008.safetensors",
|
| 432 |
+
"language_model.model.layers.2.mlp.up_proj.weight": "model-00001-of-00008.safetensors",
|
| 433 |
+
"language_model.model.layers.2.post_attention_layernorm.weight": "model-00001-of-00008.safetensors",
|
| 434 |
+
"language_model.model.layers.2.self_attn.k_proj.bias": "model-00001-of-00008.safetensors",
|
| 435 |
+
"language_model.model.layers.2.self_attn.k_proj.weight": "model-00001-of-00008.safetensors",
|
| 436 |
+
"language_model.model.layers.2.self_attn.o_proj.weight": "model-00001-of-00008.safetensors",
|
| 437 |
+
"language_model.model.layers.2.self_attn.q_proj.bias": "model-00001-of-00008.safetensors",
|
| 438 |
+
"language_model.model.layers.2.self_attn.q_proj.weight": "model-00001-of-00008.safetensors",
|
| 439 |
+
"language_model.model.layers.2.self_attn.v_proj.bias": "model-00001-of-00008.safetensors",
|
| 440 |
+
"language_model.model.layers.2.self_attn.v_proj.weight": "model-00001-of-00008.safetensors",
|
| 441 |
+
"language_model.model.layers.20.input_layernorm.weight": "model-00005-of-00008.safetensors",
|
| 442 |
+
"language_model.model.layers.20.mlp.down_proj.weight": "model-00005-of-00008.safetensors",
|
| 443 |
+
"language_model.model.layers.20.mlp.gate_proj.weight": "model-00005-of-00008.safetensors",
|
| 444 |
+
"language_model.model.layers.20.mlp.up_proj.weight": "model-00005-of-00008.safetensors",
|
| 445 |
+
"language_model.model.layers.20.post_attention_layernorm.weight": "model-00005-of-00008.safetensors",
|
| 446 |
+
"language_model.model.layers.20.self_attn.k_proj.bias": "model-00005-of-00008.safetensors",
|
| 447 |
+
"language_model.model.layers.20.self_attn.k_proj.weight": "model-00005-of-00008.safetensors",
|
| 448 |
+
"language_model.model.layers.20.self_attn.o_proj.weight": "model-00005-of-00008.safetensors",
|
| 449 |
+
"language_model.model.layers.20.self_attn.q_proj.bias": "model-00005-of-00008.safetensors",
|
| 450 |
+
"language_model.model.layers.20.self_attn.q_proj.weight": "model-00005-of-00008.safetensors",
|
| 451 |
+
"language_model.model.layers.20.self_attn.v_proj.bias": "model-00005-of-00008.safetensors",
|
| 452 |
+
"language_model.model.layers.20.self_attn.v_proj.weight": "model-00005-of-00008.safetensors",
|
| 453 |
+
"language_model.model.layers.21.input_layernorm.weight": "model-00005-of-00008.safetensors",
|
| 454 |
+
"language_model.model.layers.21.mlp.down_proj.weight": "model-00005-of-00008.safetensors",
|
| 455 |
+
"language_model.model.layers.21.mlp.gate_proj.weight": "model-00005-of-00008.safetensors",
|
| 456 |
+
"language_model.model.layers.21.mlp.up_proj.weight": "model-00005-of-00008.safetensors",
|
| 457 |
+
"language_model.model.layers.21.post_attention_layernorm.weight": "model-00005-of-00008.safetensors",
|
| 458 |
+
"language_model.model.layers.21.self_attn.k_proj.bias": "model-00005-of-00008.safetensors",
|
| 459 |
+
"language_model.model.layers.21.self_attn.k_proj.weight": "model-00005-of-00008.safetensors",
|
| 460 |
+
"language_model.model.layers.21.self_attn.o_proj.weight": "model-00005-of-00008.safetensors",
|
| 461 |
+
"language_model.model.layers.21.self_attn.q_proj.bias": "model-00005-of-00008.safetensors",
|
| 462 |
+
"language_model.model.layers.21.self_attn.q_proj.weight": "model-00005-of-00008.safetensors",
|
| 463 |
+
"language_model.model.layers.21.self_attn.v_proj.bias": "model-00005-of-00008.safetensors",
|
| 464 |
+
"language_model.model.layers.21.self_attn.v_proj.weight": "model-00005-of-00008.safetensors",
|
| 465 |
+
"language_model.model.layers.22.input_layernorm.weight": "model-00005-of-00008.safetensors",
|
| 466 |
+
"language_model.model.layers.22.mlp.down_proj.weight": "model-00005-of-00008.safetensors",
|
| 467 |
+
"language_model.model.layers.22.mlp.gate_proj.weight": "model-00005-of-00008.safetensors",
|
| 468 |
+
"language_model.model.layers.22.mlp.up_proj.weight": "model-00005-of-00008.safetensors",
|
| 469 |
+
"language_model.model.layers.22.post_attention_layernorm.weight": "model-00005-of-00008.safetensors",
|
| 470 |
+
"language_model.model.layers.22.self_attn.k_proj.bias": "model-00005-of-00008.safetensors",
|
| 471 |
+
"language_model.model.layers.22.self_attn.k_proj.weight": "model-00005-of-00008.safetensors",
|
| 472 |
+
"language_model.model.layers.22.self_attn.o_proj.weight": "model-00005-of-00008.safetensors",
|
| 473 |
+
"language_model.model.layers.22.self_attn.q_proj.bias": "model-00005-of-00008.safetensors",
|
| 474 |
+
"language_model.model.layers.22.self_attn.q_proj.weight": "model-00005-of-00008.safetensors",
|
| 475 |
+
"language_model.model.layers.22.self_attn.v_proj.bias": "model-00005-of-00008.safetensors",
|
| 476 |
+
"language_model.model.layers.22.self_attn.v_proj.weight": "model-00005-of-00008.safetensors",
|
| 477 |
+
"language_model.model.layers.23.input_layernorm.weight": "model-00005-of-00008.safetensors",
|
| 478 |
+
"language_model.model.layers.23.mlp.down_proj.weight": "model-00005-of-00008.safetensors",
|
| 479 |
+
"language_model.model.layers.23.mlp.gate_proj.weight": "model-00005-of-00008.safetensors",
|
| 480 |
+
"language_model.model.layers.23.mlp.up_proj.weight": "model-00005-of-00008.safetensors",
|
| 481 |
+
"language_model.model.layers.23.post_attention_layernorm.weight": "model-00005-of-00008.safetensors",
|
| 482 |
+
"language_model.model.layers.23.self_attn.k_proj.bias": "model-00005-of-00008.safetensors",
|
| 483 |
+
"language_model.model.layers.23.self_attn.k_proj.weight": "model-00005-of-00008.safetensors",
|
| 484 |
+
"language_model.model.layers.23.self_attn.o_proj.weight": "model-00005-of-00008.safetensors",
|
| 485 |
+
"language_model.model.layers.23.self_attn.q_proj.bias": "model-00005-of-00008.safetensors",
|
| 486 |
+
"language_model.model.layers.23.self_attn.q_proj.weight": "model-00005-of-00008.safetensors",
|
| 487 |
+
"language_model.model.layers.23.self_attn.v_proj.bias": "model-00005-of-00008.safetensors",
|
| 488 |
+
"language_model.model.layers.23.self_attn.v_proj.weight": "model-00005-of-00008.safetensors",
|
| 489 |
+
"language_model.model.layers.24.input_layernorm.weight": "model-00006-of-00008.safetensors",
|
| 490 |
+
"language_model.model.layers.24.mlp.down_proj.weight": "model-00006-of-00008.safetensors",
|
| 491 |
+
"language_model.model.layers.24.mlp.gate_proj.weight": "model-00006-of-00008.safetensors",
|
| 492 |
+
"language_model.model.layers.24.mlp.up_proj.weight": "model-00006-of-00008.safetensors",
|
| 493 |
+
"language_model.model.layers.24.post_attention_layernorm.weight": "model-00006-of-00008.safetensors",
|
| 494 |
+
"language_model.model.layers.24.self_attn.k_proj.bias": "model-00005-of-00008.safetensors",
|
| 495 |
+
"language_model.model.layers.24.self_attn.k_proj.weight": "model-00005-of-00008.safetensors",
|
| 496 |
+
"language_model.model.layers.24.self_attn.o_proj.weight": "model-00006-of-00008.safetensors",
|
| 497 |
+
"language_model.model.layers.24.self_attn.q_proj.bias": "model-00005-of-00008.safetensors",
|
| 498 |
+
"language_model.model.layers.24.self_attn.q_proj.weight": "model-00005-of-00008.safetensors",
|
| 499 |
+
"language_model.model.layers.24.self_attn.v_proj.bias": "model-00005-of-00008.safetensors",
|
| 500 |
+
"language_model.model.layers.24.self_attn.v_proj.weight": "model-00005-of-00008.safetensors",
|
| 501 |
+
"language_model.model.layers.25.input_layernorm.weight": "model-00006-of-00008.safetensors",
|
| 502 |
+
"language_model.model.layers.25.mlp.down_proj.weight": "model-00006-of-00008.safetensors",
|
| 503 |
+
"language_model.model.layers.25.mlp.gate_proj.weight": "model-00006-of-00008.safetensors",
|
| 504 |
+
"language_model.model.layers.25.mlp.up_proj.weight": "model-00006-of-00008.safetensors",
|
| 505 |
+
"language_model.model.layers.25.post_attention_layernorm.weight": "model-00006-of-00008.safetensors",
|
| 506 |
+
"language_model.model.layers.25.self_attn.k_proj.bias": "model-00006-of-00008.safetensors",
|
| 507 |
+
"language_model.model.layers.25.self_attn.k_proj.weight": "model-00006-of-00008.safetensors",
|
| 508 |
+
"language_model.model.layers.25.self_attn.o_proj.weight": "model-00006-of-00008.safetensors",
|
| 509 |
+
"language_model.model.layers.25.self_attn.q_proj.bias": "model-00006-of-00008.safetensors",
|
| 510 |
+
"language_model.model.layers.25.self_attn.q_proj.weight": "model-00006-of-00008.safetensors",
|
| 511 |
+
"language_model.model.layers.25.self_attn.v_proj.bias": "model-00006-of-00008.safetensors",
|
| 512 |
+
"language_model.model.layers.25.self_attn.v_proj.weight": "model-00006-of-00008.safetensors",
|
| 513 |
+
"language_model.model.layers.26.input_layernorm.weight": "model-00006-of-00008.safetensors",
|
| 514 |
+
"language_model.model.layers.26.mlp.down_proj.weight": "model-00006-of-00008.safetensors",
|
| 515 |
+
"language_model.model.layers.26.mlp.gate_proj.weight": "model-00006-of-00008.safetensors",
|
| 516 |
+
"language_model.model.layers.26.mlp.up_proj.weight": "model-00006-of-00008.safetensors",
|
| 517 |
+
"language_model.model.layers.26.post_attention_layernorm.weight": "model-00006-of-00008.safetensors",
|
| 518 |
+
"language_model.model.layers.26.self_attn.k_proj.bias": "model-00006-of-00008.safetensors",
|
| 519 |
+
"language_model.model.layers.26.self_attn.k_proj.weight": "model-00006-of-00008.safetensors",
|
| 520 |
+
"language_model.model.layers.26.self_attn.o_proj.weight": "model-00006-of-00008.safetensors",
|
| 521 |
+
"language_model.model.layers.26.self_attn.q_proj.bias": "model-00006-of-00008.safetensors",
|
| 522 |
+
"language_model.model.layers.26.self_attn.q_proj.weight": "model-00006-of-00008.safetensors",
|
| 523 |
+
"language_model.model.layers.26.self_attn.v_proj.bias": "model-00006-of-00008.safetensors",
|
| 524 |
+
"language_model.model.layers.26.self_attn.v_proj.weight": "model-00006-of-00008.safetensors",
|
| 525 |
+
"language_model.model.layers.27.input_layernorm.weight": "model-00006-of-00008.safetensors",
|
| 526 |
+
"language_model.model.layers.27.mlp.down_proj.weight": "model-00006-of-00008.safetensors",
|
| 527 |
+
"language_model.model.layers.27.mlp.gate_proj.weight": "model-00006-of-00008.safetensors",
|
| 528 |
+
"language_model.model.layers.27.mlp.up_proj.weight": "model-00006-of-00008.safetensors",
|
| 529 |
+
"language_model.model.layers.27.post_attention_layernorm.weight": "model-00006-of-00008.safetensors",
|
| 530 |
+
"language_model.model.layers.27.self_attn.k_proj.bias": "model-00006-of-00008.safetensors",
|
| 531 |
+
"language_model.model.layers.27.self_attn.k_proj.weight": "model-00006-of-00008.safetensors",
|
| 532 |
+
"language_model.model.layers.27.self_attn.o_proj.weight": "model-00006-of-00008.safetensors",
|
| 533 |
+
"language_model.model.layers.27.self_attn.q_proj.bias": "model-00006-of-00008.safetensors",
|
| 534 |
+
"language_model.model.layers.27.self_attn.q_proj.weight": "model-00006-of-00008.safetensors",
|
| 535 |
+
"language_model.model.layers.27.self_attn.v_proj.bias": "model-00006-of-00008.safetensors",
|
| 536 |
+
"language_model.model.layers.27.self_attn.v_proj.weight": "model-00006-of-00008.safetensors",
|
| 537 |
+
"language_model.model.layers.3.input_layernorm.weight": "model-00002-of-00008.safetensors",
|
| 538 |
+
"language_model.model.layers.3.mlp.down_proj.weight": "model-00002-of-00008.safetensors",
|
| 539 |
+
"language_model.model.layers.3.mlp.gate_proj.weight": "model-00002-of-00008.safetensors",
|
| 540 |
+
"language_model.model.layers.3.mlp.up_proj.weight": "model-00002-of-00008.safetensors",
|
| 541 |
+
"language_model.model.layers.3.post_attention_layernorm.weight": "model-00002-of-00008.safetensors",
|
| 542 |
+
"language_model.model.layers.3.self_attn.k_proj.bias": "model-00002-of-00008.safetensors",
|
| 543 |
+
"language_model.model.layers.3.self_attn.k_proj.weight": "model-00002-of-00008.safetensors",
|
| 544 |
+
"language_model.model.layers.3.self_attn.o_proj.weight": "model-00002-of-00008.safetensors",
|
| 545 |
+
"language_model.model.layers.3.self_attn.q_proj.bias": "model-00002-of-00008.safetensors",
|
| 546 |
+
"language_model.model.layers.3.self_attn.q_proj.weight": "model-00002-of-00008.safetensors",
|
| 547 |
+
"language_model.model.layers.3.self_attn.v_proj.bias": "model-00002-of-00008.safetensors",
|
| 548 |
+
"language_model.model.layers.3.self_attn.v_proj.weight": "model-00002-of-00008.safetensors",
|
| 549 |
+
"language_model.model.layers.4.input_layernorm.weight": "model-00002-of-00008.safetensors",
|
| 550 |
+
"language_model.model.layers.4.mlp.down_proj.weight": "model-00002-of-00008.safetensors",
|
| 551 |
+
"language_model.model.layers.4.mlp.gate_proj.weight": "model-00002-of-00008.safetensors",
|
| 552 |
+
"language_model.model.layers.4.mlp.up_proj.weight": "model-00002-of-00008.safetensors",
|
| 553 |
+
"language_model.model.layers.4.post_attention_layernorm.weight": "model-00002-of-00008.safetensors",
|
| 554 |
+
"language_model.model.layers.4.self_attn.k_proj.bias": "model-00002-of-00008.safetensors",
|
| 555 |
+
"language_model.model.layers.4.self_attn.k_proj.weight": "model-00002-of-00008.safetensors",
|
| 556 |
+
"language_model.model.layers.4.self_attn.o_proj.weight": "model-00002-of-00008.safetensors",
|
| 557 |
+
"language_model.model.layers.4.self_attn.q_proj.bias": "model-00002-of-00008.safetensors",
|
| 558 |
+
"language_model.model.layers.4.self_attn.q_proj.weight": "model-00002-of-00008.safetensors",
|
| 559 |
+
"language_model.model.layers.4.self_attn.v_proj.bias": "model-00002-of-00008.safetensors",
|
| 560 |
+
"language_model.model.layers.4.self_attn.v_proj.weight": "model-00002-of-00008.safetensors",
|
| 561 |
+
"language_model.model.layers.5.input_layernorm.weight": "model-00002-of-00008.safetensors",
|
| 562 |
+
"language_model.model.layers.5.mlp.down_proj.weight": "model-00002-of-00008.safetensors",
|
| 563 |
+
"language_model.model.layers.5.mlp.gate_proj.weight": "model-00002-of-00008.safetensors",
|
| 564 |
+
"language_model.model.layers.5.mlp.up_proj.weight": "model-00002-of-00008.safetensors",
|
| 565 |
+
"language_model.model.layers.5.post_attention_layernorm.weight": "model-00002-of-00008.safetensors",
|
| 566 |
+
"language_model.model.layers.5.self_attn.k_proj.bias": "model-00002-of-00008.safetensors",
|
| 567 |
+
"language_model.model.layers.5.self_attn.k_proj.weight": "model-00002-of-00008.safetensors",
|
| 568 |
+
"language_model.model.layers.5.self_attn.o_proj.weight": "model-00002-of-00008.safetensors",
|
| 569 |
+
"language_model.model.layers.5.self_attn.q_proj.bias": "model-00002-of-00008.safetensors",
|
| 570 |
+
"language_model.model.layers.5.self_attn.q_proj.weight": "model-00002-of-00008.safetensors",
|
| 571 |
+
"language_model.model.layers.5.self_attn.v_proj.bias": "model-00002-of-00008.safetensors",
|
| 572 |
+
"language_model.model.layers.5.self_attn.v_proj.weight": "model-00002-of-00008.safetensors",
|
| 573 |
+
"language_model.model.layers.6.input_layernorm.weight": "model-00002-of-00008.safetensors",
|
| 574 |
+
"language_model.model.layers.6.mlp.down_proj.weight": "model-00002-of-00008.safetensors",
|
| 575 |
+
"language_model.model.layers.6.mlp.gate_proj.weight": "model-00002-of-00008.safetensors",
|
| 576 |
+
"language_model.model.layers.6.mlp.up_proj.weight": "model-00002-of-00008.safetensors",
|
| 577 |
+
"language_model.model.layers.6.post_attention_layernorm.weight": "model-00002-of-00008.safetensors",
|
| 578 |
+
"language_model.model.layers.6.self_attn.k_proj.bias": "model-00002-of-00008.safetensors",
|
| 579 |
+
"language_model.model.layers.6.self_attn.k_proj.weight": "model-00002-of-00008.safetensors",
|
| 580 |
+
"language_model.model.layers.6.self_attn.o_proj.weight": "model-00002-of-00008.safetensors",
|
| 581 |
+
"language_model.model.layers.6.self_attn.q_proj.bias": "model-00002-of-00008.safetensors",
|
| 582 |
+
"language_model.model.layers.6.self_attn.q_proj.weight": "model-00002-of-00008.safetensors",
|
| 583 |
+
"language_model.model.layers.6.self_attn.v_proj.bias": "model-00002-of-00008.safetensors",
|
| 584 |
+
"language_model.model.layers.6.self_attn.v_proj.weight": "model-00002-of-00008.safetensors",
|
| 585 |
+
"language_model.model.layers.7.input_layernorm.weight": "model-00002-of-00008.safetensors",
|
| 586 |
+
"language_model.model.layers.7.mlp.down_proj.weight": "model-00002-of-00008.safetensors",
|
| 587 |
+
"language_model.model.layers.7.mlp.gate_proj.weight": "model-00002-of-00008.safetensors",
|
| 588 |
+
"language_model.model.layers.7.mlp.up_proj.weight": "model-00002-of-00008.safetensors",
|
| 589 |
+
"language_model.model.layers.7.post_attention_layernorm.weight": "model-00002-of-00008.safetensors",
|
| 590 |
+
"language_model.model.layers.7.self_attn.k_proj.bias": "model-00002-of-00008.safetensors",
|
| 591 |
+
"language_model.model.layers.7.self_attn.k_proj.weight": "model-00002-of-00008.safetensors",
|
| 592 |
+
"language_model.model.layers.7.self_attn.o_proj.weight": "model-00002-of-00008.safetensors",
|
| 593 |
+
"language_model.model.layers.7.self_attn.q_proj.bias": "model-00002-of-00008.safetensors",
|
| 594 |
+
"language_model.model.layers.7.self_attn.q_proj.weight": "model-00002-of-00008.safetensors",
|
| 595 |
+
"language_model.model.layers.7.self_attn.v_proj.bias": "model-00002-of-00008.safetensors",
|
| 596 |
+
"language_model.model.layers.7.self_attn.v_proj.weight": "model-00002-of-00008.safetensors",
|
| 597 |
+
"language_model.model.layers.8.input_layernorm.weight": "model-00003-of-00008.safetensors",
|
| 598 |
+
"language_model.model.layers.8.mlp.down_proj.weight": "model-00003-of-00008.safetensors",
|
| 599 |
+
"language_model.model.layers.8.mlp.gate_proj.weight": "model-00003-of-00008.safetensors",
|
| 600 |
+
"language_model.model.layers.8.mlp.up_proj.weight": "model-00003-of-00008.safetensors",
|
| 601 |
+
"language_model.model.layers.8.post_attention_layernorm.weight": "model-00003-of-00008.safetensors",
|
| 602 |
+
"language_model.model.layers.8.self_attn.k_proj.bias": "model-00002-of-00008.safetensors",
|
| 603 |
+
"language_model.model.layers.8.self_attn.k_proj.weight": "model-00002-of-00008.safetensors",
|
| 604 |
+
"language_model.model.layers.8.self_attn.o_proj.weight": "model-00002-of-00008.safetensors",
|
| 605 |
+
"language_model.model.layers.8.self_attn.q_proj.bias": "model-00002-of-00008.safetensors",
|
| 606 |
+
"language_model.model.layers.8.self_attn.q_proj.weight": "model-00002-of-00008.safetensors",
|
| 607 |
+
"language_model.model.layers.8.self_attn.v_proj.bias": "model-00002-of-00008.safetensors",
|
| 608 |
+
"language_model.model.layers.8.self_attn.v_proj.weight": "model-00002-of-00008.safetensors",
|
| 609 |
+
"language_model.model.layers.9.input_layernorm.weight": "model-00003-of-00008.safetensors",
|
| 610 |
+
"language_model.model.layers.9.mlp.down_proj.weight": "model-00003-of-00008.safetensors",
|
| 611 |
+
"language_model.model.layers.9.mlp.gate_proj.weight": "model-00003-of-00008.safetensors",
|
| 612 |
+
"language_model.model.layers.9.mlp.up_proj.weight": "model-00003-of-00008.safetensors",
|
| 613 |
+
"language_model.model.layers.9.post_attention_layernorm.weight": "model-00003-of-00008.safetensors",
|
| 614 |
+
"language_model.model.layers.9.self_attn.k_proj.bias": "model-00003-of-00008.safetensors",
|
| 615 |
+
"language_model.model.layers.9.self_attn.k_proj.weight": "model-00003-of-00008.safetensors",
|
| 616 |
+
"language_model.model.layers.9.self_attn.o_proj.weight": "model-00003-of-00008.safetensors",
|
| 617 |
+
"language_model.model.layers.9.self_attn.q_proj.bias": "model-00003-of-00008.safetensors",
|
| 618 |
+
"language_model.model.layers.9.self_attn.q_proj.weight": "model-00003-of-00008.safetensors",
|
| 619 |
+
"language_model.model.layers.9.self_attn.v_proj.bias": "model-00003-of-00008.safetensors",
|
| 620 |
+
"language_model.model.layers.9.self_attn.v_proj.weight": "model-00003-of-00008.safetensors",
|
| 621 |
+
"language_model.model.norm.weight": "model-00006-of-00008.safetensors",
|
| 622 |
+
"multi_modal_projector.acoustic_linear_1.bias": "model-00007-of-00008.safetensors",
|
| 623 |
+
"multi_modal_projector.acoustic_linear_1.weight": "model-00007-of-00008.safetensors",
|
| 624 |
+
"multi_modal_projector.acoustic_linear_2.bias": "model-00007-of-00008.safetensors",
|
| 625 |
+
"multi_modal_projector.acoustic_linear_2.weight": "model-00007-of-00008.safetensors",
|
| 626 |
+
"multi_modal_projector.acoustic_norm.weight": "model-00007-of-00008.safetensors",
|
| 627 |
+
"multi_modal_projector.semantic_linear_1.bias": "model-00007-of-00008.safetensors",
|
| 628 |
+
"multi_modal_projector.semantic_linear_1.weight": "model-00007-of-00008.safetensors",
|
| 629 |
+
"multi_modal_projector.semantic_linear_2.bias": "model-00007-of-00008.safetensors",
|
| 630 |
+
"multi_modal_projector.semantic_linear_2.weight": "model-00007-of-00008.safetensors",
|
| 631 |
+
"multi_modal_projector.semantic_norm.weight": "model-00007-of-00008.safetensors",
|
| 632 |
+
"semantic_tokenizer_encoder.conv_layers.0.conv.conv.bias": "model-00007-of-00008.safetensors",
|
| 633 |
+
"semantic_tokenizer_encoder.conv_layers.0.conv.conv.weight": "model-00007-of-00008.safetensors",
|
| 634 |
+
"semantic_tokenizer_encoder.conv_layers.0.stage.0.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 635 |
+
"semantic_tokenizer_encoder.conv_layers.0.stage.0.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 636 |
+
"semantic_tokenizer_encoder.conv_layers.0.stage.0.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 637 |
+
"semantic_tokenizer_encoder.conv_layers.0.stage.0.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 638 |
+
"semantic_tokenizer_encoder.conv_layers.0.stage.0.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 639 |
+
"semantic_tokenizer_encoder.conv_layers.0.stage.0.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 640 |
+
"semantic_tokenizer_encoder.conv_layers.0.stage.0.gamma": "model-00007-of-00008.safetensors",
|
| 641 |
+
"semantic_tokenizer_encoder.conv_layers.0.stage.0.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 642 |
+
"semantic_tokenizer_encoder.conv_layers.0.stage.0.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 643 |
+
"semantic_tokenizer_encoder.conv_layers.0.stage.0.norm.weight": "model-00007-of-00008.safetensors",
|
| 644 |
+
"semantic_tokenizer_encoder.conv_layers.0.stage.1.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 645 |
+
"semantic_tokenizer_encoder.conv_layers.0.stage.1.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 646 |
+
"semantic_tokenizer_encoder.conv_layers.0.stage.1.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 647 |
+
"semantic_tokenizer_encoder.conv_layers.0.stage.1.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 648 |
+
"semantic_tokenizer_encoder.conv_layers.0.stage.1.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 649 |
+
"semantic_tokenizer_encoder.conv_layers.0.stage.1.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 650 |
+
"semantic_tokenizer_encoder.conv_layers.0.stage.1.gamma": "model-00007-of-00008.safetensors",
|
| 651 |
+
"semantic_tokenizer_encoder.conv_layers.0.stage.1.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 652 |
+
"semantic_tokenizer_encoder.conv_layers.0.stage.1.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 653 |
+
"semantic_tokenizer_encoder.conv_layers.0.stage.1.norm.weight": "model-00007-of-00008.safetensors",
|
| 654 |
+
"semantic_tokenizer_encoder.conv_layers.0.stage.2.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 655 |
+
"semantic_tokenizer_encoder.conv_layers.0.stage.2.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 656 |
+
"semantic_tokenizer_encoder.conv_layers.0.stage.2.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 657 |
+
"semantic_tokenizer_encoder.conv_layers.0.stage.2.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 658 |
+
"semantic_tokenizer_encoder.conv_layers.0.stage.2.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 659 |
+
"semantic_tokenizer_encoder.conv_layers.0.stage.2.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 660 |
+
"semantic_tokenizer_encoder.conv_layers.0.stage.2.gamma": "model-00007-of-00008.safetensors",
|
| 661 |
+
"semantic_tokenizer_encoder.conv_layers.0.stage.2.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 662 |
+
"semantic_tokenizer_encoder.conv_layers.0.stage.2.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 663 |
+
"semantic_tokenizer_encoder.conv_layers.0.stage.2.norm.weight": "model-00007-of-00008.safetensors",
|
| 664 |
+
"semantic_tokenizer_encoder.conv_layers.1.conv.conv.bias": "model-00007-of-00008.safetensors",
|
| 665 |
+
"semantic_tokenizer_encoder.conv_layers.1.conv.conv.weight": "model-00007-of-00008.safetensors",
|
| 666 |
+
"semantic_tokenizer_encoder.conv_layers.1.stage.0.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 667 |
+
"semantic_tokenizer_encoder.conv_layers.1.stage.0.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 668 |
+
"semantic_tokenizer_encoder.conv_layers.1.stage.0.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 669 |
+
"semantic_tokenizer_encoder.conv_layers.1.stage.0.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 670 |
+
"semantic_tokenizer_encoder.conv_layers.1.stage.0.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 671 |
+
"semantic_tokenizer_encoder.conv_layers.1.stage.0.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 672 |
+
"semantic_tokenizer_encoder.conv_layers.1.stage.0.gamma": "model-00007-of-00008.safetensors",
|
| 673 |
+
"semantic_tokenizer_encoder.conv_layers.1.stage.0.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 674 |
+
"semantic_tokenizer_encoder.conv_layers.1.stage.0.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 675 |
+
"semantic_tokenizer_encoder.conv_layers.1.stage.0.norm.weight": "model-00007-of-00008.safetensors",
|
| 676 |
+
"semantic_tokenizer_encoder.conv_layers.1.stage.1.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 677 |
+
"semantic_tokenizer_encoder.conv_layers.1.stage.1.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 678 |
+
"semantic_tokenizer_encoder.conv_layers.1.stage.1.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 679 |
+
"semantic_tokenizer_encoder.conv_layers.1.stage.1.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 680 |
+
"semantic_tokenizer_encoder.conv_layers.1.stage.1.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 681 |
+
"semantic_tokenizer_encoder.conv_layers.1.stage.1.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 682 |
+
"semantic_tokenizer_encoder.conv_layers.1.stage.1.gamma": "model-00007-of-00008.safetensors",
|
| 683 |
+
"semantic_tokenizer_encoder.conv_layers.1.stage.1.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 684 |
+
"semantic_tokenizer_encoder.conv_layers.1.stage.1.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 685 |
+
"semantic_tokenizer_encoder.conv_layers.1.stage.1.norm.weight": "model-00007-of-00008.safetensors",
|
| 686 |
+
"semantic_tokenizer_encoder.conv_layers.1.stage.2.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 687 |
+
"semantic_tokenizer_encoder.conv_layers.1.stage.2.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 688 |
+
"semantic_tokenizer_encoder.conv_layers.1.stage.2.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 689 |
+
"semantic_tokenizer_encoder.conv_layers.1.stage.2.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 690 |
+
"semantic_tokenizer_encoder.conv_layers.1.stage.2.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 691 |
+
"semantic_tokenizer_encoder.conv_layers.1.stage.2.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 692 |
+
"semantic_tokenizer_encoder.conv_layers.1.stage.2.gamma": "model-00007-of-00008.safetensors",
|
| 693 |
+
"semantic_tokenizer_encoder.conv_layers.1.stage.2.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 694 |
+
"semantic_tokenizer_encoder.conv_layers.1.stage.2.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 695 |
+
"semantic_tokenizer_encoder.conv_layers.1.stage.2.norm.weight": "model-00007-of-00008.safetensors",
|
| 696 |
+
"semantic_tokenizer_encoder.conv_layers.2.conv.conv.bias": "model-00007-of-00008.safetensors",
|
| 697 |
+
"semantic_tokenizer_encoder.conv_layers.2.conv.conv.weight": "model-00007-of-00008.safetensors",
|
| 698 |
+
"semantic_tokenizer_encoder.conv_layers.2.stage.0.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 699 |
+
"semantic_tokenizer_encoder.conv_layers.2.stage.0.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 700 |
+
"semantic_tokenizer_encoder.conv_layers.2.stage.0.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 701 |
+
"semantic_tokenizer_encoder.conv_layers.2.stage.0.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 702 |
+
"semantic_tokenizer_encoder.conv_layers.2.stage.0.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 703 |
+
"semantic_tokenizer_encoder.conv_layers.2.stage.0.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 704 |
+
"semantic_tokenizer_encoder.conv_layers.2.stage.0.gamma": "model-00007-of-00008.safetensors",
|
| 705 |
+
"semantic_tokenizer_encoder.conv_layers.2.stage.0.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 706 |
+
"semantic_tokenizer_encoder.conv_layers.2.stage.0.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 707 |
+
"semantic_tokenizer_encoder.conv_layers.2.stage.0.norm.weight": "model-00007-of-00008.safetensors",
|
| 708 |
+
"semantic_tokenizer_encoder.conv_layers.2.stage.1.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 709 |
+
"semantic_tokenizer_encoder.conv_layers.2.stage.1.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 710 |
+
"semantic_tokenizer_encoder.conv_layers.2.stage.1.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 711 |
+
"semantic_tokenizer_encoder.conv_layers.2.stage.1.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 712 |
+
"semantic_tokenizer_encoder.conv_layers.2.stage.1.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 713 |
+
"semantic_tokenizer_encoder.conv_layers.2.stage.1.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 714 |
+
"semantic_tokenizer_encoder.conv_layers.2.stage.1.gamma": "model-00007-of-00008.safetensors",
|
| 715 |
+
"semantic_tokenizer_encoder.conv_layers.2.stage.1.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 716 |
+
"semantic_tokenizer_encoder.conv_layers.2.stage.1.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 717 |
+
"semantic_tokenizer_encoder.conv_layers.2.stage.1.norm.weight": "model-00007-of-00008.safetensors",
|
| 718 |
+
"semantic_tokenizer_encoder.conv_layers.2.stage.2.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 719 |
+
"semantic_tokenizer_encoder.conv_layers.2.stage.2.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 720 |
+
"semantic_tokenizer_encoder.conv_layers.2.stage.2.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 721 |
+
"semantic_tokenizer_encoder.conv_layers.2.stage.2.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 722 |
+
"semantic_tokenizer_encoder.conv_layers.2.stage.2.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 723 |
+
"semantic_tokenizer_encoder.conv_layers.2.stage.2.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 724 |
+
"semantic_tokenizer_encoder.conv_layers.2.stage.2.gamma": "model-00007-of-00008.safetensors",
|
| 725 |
+
"semantic_tokenizer_encoder.conv_layers.2.stage.2.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 726 |
+
"semantic_tokenizer_encoder.conv_layers.2.stage.2.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 727 |
+
"semantic_tokenizer_encoder.conv_layers.2.stage.2.norm.weight": "model-00007-of-00008.safetensors",
|
| 728 |
+
"semantic_tokenizer_encoder.conv_layers.3.conv.conv.bias": "model-00007-of-00008.safetensors",
|
| 729 |
+
"semantic_tokenizer_encoder.conv_layers.3.conv.conv.weight": "model-00007-of-00008.safetensors",
|
| 730 |
+
"semantic_tokenizer_encoder.conv_layers.3.stage.0.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 731 |
+
"semantic_tokenizer_encoder.conv_layers.3.stage.0.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 732 |
+
"semantic_tokenizer_encoder.conv_layers.3.stage.0.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 733 |
+
"semantic_tokenizer_encoder.conv_layers.3.stage.0.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 734 |
+
"semantic_tokenizer_encoder.conv_layers.3.stage.0.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 735 |
+
"semantic_tokenizer_encoder.conv_layers.3.stage.0.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 736 |
+
"semantic_tokenizer_encoder.conv_layers.3.stage.0.gamma": "model-00007-of-00008.safetensors",
|
| 737 |
+
"semantic_tokenizer_encoder.conv_layers.3.stage.0.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 738 |
+
"semantic_tokenizer_encoder.conv_layers.3.stage.0.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 739 |
+
"semantic_tokenizer_encoder.conv_layers.3.stage.0.norm.weight": "model-00007-of-00008.safetensors",
|
| 740 |
+
"semantic_tokenizer_encoder.conv_layers.3.stage.1.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 741 |
+
"semantic_tokenizer_encoder.conv_layers.3.stage.1.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 742 |
+
"semantic_tokenizer_encoder.conv_layers.3.stage.1.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 743 |
+
"semantic_tokenizer_encoder.conv_layers.3.stage.1.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 744 |
+
"semantic_tokenizer_encoder.conv_layers.3.stage.1.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 745 |
+
"semantic_tokenizer_encoder.conv_layers.3.stage.1.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 746 |
+
"semantic_tokenizer_encoder.conv_layers.3.stage.1.gamma": "model-00007-of-00008.safetensors",
|
| 747 |
+
"semantic_tokenizer_encoder.conv_layers.3.stage.1.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 748 |
+
"semantic_tokenizer_encoder.conv_layers.3.stage.1.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 749 |
+
"semantic_tokenizer_encoder.conv_layers.3.stage.1.norm.weight": "model-00007-of-00008.safetensors",
|
| 750 |
+
"semantic_tokenizer_encoder.conv_layers.3.stage.2.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 751 |
+
"semantic_tokenizer_encoder.conv_layers.3.stage.2.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 752 |
+
"semantic_tokenizer_encoder.conv_layers.3.stage.2.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 753 |
+
"semantic_tokenizer_encoder.conv_layers.3.stage.2.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 754 |
+
"semantic_tokenizer_encoder.conv_layers.3.stage.2.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 755 |
+
"semantic_tokenizer_encoder.conv_layers.3.stage.2.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 756 |
+
"semantic_tokenizer_encoder.conv_layers.3.stage.2.gamma": "model-00007-of-00008.safetensors",
|
| 757 |
+
"semantic_tokenizer_encoder.conv_layers.3.stage.2.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 758 |
+
"semantic_tokenizer_encoder.conv_layers.3.stage.2.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 759 |
+
"semantic_tokenizer_encoder.conv_layers.3.stage.2.norm.weight": "model-00007-of-00008.safetensors",
|
| 760 |
+
"semantic_tokenizer_encoder.conv_layers.4.conv.conv.bias": "model-00007-of-00008.safetensors",
|
| 761 |
+
"semantic_tokenizer_encoder.conv_layers.4.conv.conv.weight": "model-00007-of-00008.safetensors",
|
| 762 |
+
"semantic_tokenizer_encoder.conv_layers.4.stage.0.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 763 |
+
"semantic_tokenizer_encoder.conv_layers.4.stage.0.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 764 |
+
"semantic_tokenizer_encoder.conv_layers.4.stage.0.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 765 |
+
"semantic_tokenizer_encoder.conv_layers.4.stage.0.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 766 |
+
"semantic_tokenizer_encoder.conv_layers.4.stage.0.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 767 |
+
"semantic_tokenizer_encoder.conv_layers.4.stage.0.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 768 |
+
"semantic_tokenizer_encoder.conv_layers.4.stage.0.gamma": "model-00007-of-00008.safetensors",
|
| 769 |
+
"semantic_tokenizer_encoder.conv_layers.4.stage.0.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 770 |
+
"semantic_tokenizer_encoder.conv_layers.4.stage.0.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 771 |
+
"semantic_tokenizer_encoder.conv_layers.4.stage.0.norm.weight": "model-00007-of-00008.safetensors",
|
| 772 |
+
"semantic_tokenizer_encoder.conv_layers.4.stage.1.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 773 |
+
"semantic_tokenizer_encoder.conv_layers.4.stage.1.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 774 |
+
"semantic_tokenizer_encoder.conv_layers.4.stage.1.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 775 |
+
"semantic_tokenizer_encoder.conv_layers.4.stage.1.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 776 |
+
"semantic_tokenizer_encoder.conv_layers.4.stage.1.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 777 |
+
"semantic_tokenizer_encoder.conv_layers.4.stage.1.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 778 |
+
"semantic_tokenizer_encoder.conv_layers.4.stage.1.gamma": "model-00007-of-00008.safetensors",
|
| 779 |
+
"semantic_tokenizer_encoder.conv_layers.4.stage.1.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 780 |
+
"semantic_tokenizer_encoder.conv_layers.4.stage.1.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 781 |
+
"semantic_tokenizer_encoder.conv_layers.4.stage.1.norm.weight": "model-00007-of-00008.safetensors",
|
| 782 |
+
"semantic_tokenizer_encoder.conv_layers.4.stage.2.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 783 |
+
"semantic_tokenizer_encoder.conv_layers.4.stage.2.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 784 |
+
"semantic_tokenizer_encoder.conv_layers.4.stage.2.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 785 |
+
"semantic_tokenizer_encoder.conv_layers.4.stage.2.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 786 |
+
"semantic_tokenizer_encoder.conv_layers.4.stage.2.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 787 |
+
"semantic_tokenizer_encoder.conv_layers.4.stage.2.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 788 |
+
"semantic_tokenizer_encoder.conv_layers.4.stage.2.gamma": "model-00007-of-00008.safetensors",
|
| 789 |
+
"semantic_tokenizer_encoder.conv_layers.4.stage.2.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 790 |
+
"semantic_tokenizer_encoder.conv_layers.4.stage.2.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 791 |
+
"semantic_tokenizer_encoder.conv_layers.4.stage.2.norm.weight": "model-00007-of-00008.safetensors",
|
| 792 |
+
"semantic_tokenizer_encoder.conv_layers.5.conv.conv.bias": "model-00007-of-00008.safetensors",
|
| 793 |
+
"semantic_tokenizer_encoder.conv_layers.5.conv.conv.weight": "model-00007-of-00008.safetensors",
|
| 794 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.0.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 795 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.0.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 796 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.0.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 797 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.0.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 798 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.0.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 799 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.0.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 800 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.0.gamma": "model-00007-of-00008.safetensors",
|
| 801 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.0.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 802 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.0.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 803 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.0.norm.weight": "model-00007-of-00008.safetensors",
|
| 804 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.1.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 805 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.1.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 806 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.1.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 807 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.1.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 808 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.1.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 809 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.1.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 810 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.1.gamma": "model-00007-of-00008.safetensors",
|
| 811 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.1.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 812 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.1.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 813 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.1.norm.weight": "model-00007-of-00008.safetensors",
|
| 814 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.2.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 815 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.2.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 816 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.2.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 817 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.2.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 818 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.2.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 819 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.2.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 820 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.2.gamma": "model-00007-of-00008.safetensors",
|
| 821 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.2.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 822 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.2.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 823 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.2.norm.weight": "model-00007-of-00008.safetensors",
|
| 824 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.3.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 825 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.3.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 826 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.3.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 827 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.3.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 828 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.3.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 829 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.3.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 830 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.3.gamma": "model-00007-of-00008.safetensors",
|
| 831 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.3.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 832 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.3.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 833 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.3.norm.weight": "model-00007-of-00008.safetensors",
|
| 834 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.4.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 835 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.4.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 836 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.4.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 837 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.4.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 838 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.4.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 839 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.4.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 840 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.4.gamma": "model-00007-of-00008.safetensors",
|
| 841 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.4.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 842 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.4.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 843 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.4.norm.weight": "model-00007-of-00008.safetensors",
|
| 844 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.5.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 845 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.5.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 846 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.5.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 847 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.5.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 848 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.5.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 849 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.5.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 850 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.5.gamma": "model-00007-of-00008.safetensors",
|
| 851 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.5.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 852 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.5.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 853 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.5.norm.weight": "model-00007-of-00008.safetensors",
|
| 854 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.6.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 855 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.6.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 856 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.6.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 857 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.6.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 858 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.6.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 859 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.6.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 860 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.6.gamma": "model-00007-of-00008.safetensors",
|
| 861 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.6.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 862 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.6.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 863 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.6.norm.weight": "model-00007-of-00008.safetensors",
|
| 864 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.7.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 865 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.7.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 866 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.7.ffn.linear2.bias": "model-00008-of-00008.safetensors",
|
| 867 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.7.ffn.linear2.weight": "model-00008-of-00008.safetensors",
|
| 868 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.7.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 869 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.7.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 870 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.7.gamma": "model-00007-of-00008.safetensors",
|
| 871 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.7.mixer.conv.bias": "model-00008-of-00008.safetensors",
|
| 872 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.7.mixer.conv.weight": "model-00008-of-00008.safetensors",
|
| 873 |
+
"semantic_tokenizer_encoder.conv_layers.5.stage.7.norm.weight": "model-00007-of-00008.safetensors",
|
| 874 |
+
"semantic_tokenizer_encoder.head.conv.bias": "model-00008-of-00008.safetensors",
|
| 875 |
+
"semantic_tokenizer_encoder.head.conv.weight": "model-00008-of-00008.safetensors",
|
| 876 |
+
"semantic_tokenizer_encoder.stem.conv.conv.bias": "model-00007-of-00008.safetensors",
|
| 877 |
+
"semantic_tokenizer_encoder.stem.conv.conv.weight": "model-00007-of-00008.safetensors",
|
| 878 |
+
"semantic_tokenizer_encoder.stem.stage.0.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 879 |
+
"semantic_tokenizer_encoder.stem.stage.0.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 880 |
+
"semantic_tokenizer_encoder.stem.stage.0.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 881 |
+
"semantic_tokenizer_encoder.stem.stage.0.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 882 |
+
"semantic_tokenizer_encoder.stem.stage.0.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 883 |
+
"semantic_tokenizer_encoder.stem.stage.0.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 884 |
+
"semantic_tokenizer_encoder.stem.stage.0.gamma": "model-00007-of-00008.safetensors",
|
| 885 |
+
"semantic_tokenizer_encoder.stem.stage.0.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 886 |
+
"semantic_tokenizer_encoder.stem.stage.0.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 887 |
+
"semantic_tokenizer_encoder.stem.stage.0.norm.weight": "model-00007-of-00008.safetensors",
|
| 888 |
+
"semantic_tokenizer_encoder.stem.stage.1.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 889 |
+
"semantic_tokenizer_encoder.stem.stage.1.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 890 |
+
"semantic_tokenizer_encoder.stem.stage.1.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 891 |
+
"semantic_tokenizer_encoder.stem.stage.1.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 892 |
+
"semantic_tokenizer_encoder.stem.stage.1.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 893 |
+
"semantic_tokenizer_encoder.stem.stage.1.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 894 |
+
"semantic_tokenizer_encoder.stem.stage.1.gamma": "model-00007-of-00008.safetensors",
|
| 895 |
+
"semantic_tokenizer_encoder.stem.stage.1.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 896 |
+
"semantic_tokenizer_encoder.stem.stage.1.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 897 |
+
"semantic_tokenizer_encoder.stem.stage.1.norm.weight": "model-00007-of-00008.safetensors",
|
| 898 |
+
"semantic_tokenizer_encoder.stem.stage.2.ffn.linear1.bias": "model-00007-of-00008.safetensors",
|
| 899 |
+
"semantic_tokenizer_encoder.stem.stage.2.ffn.linear1.weight": "model-00007-of-00008.safetensors",
|
| 900 |
+
"semantic_tokenizer_encoder.stem.stage.2.ffn.linear2.bias": "model-00007-of-00008.safetensors",
|
| 901 |
+
"semantic_tokenizer_encoder.stem.stage.2.ffn.linear2.weight": "model-00007-of-00008.safetensors",
|
| 902 |
+
"semantic_tokenizer_encoder.stem.stage.2.ffn_gamma": "model-00007-of-00008.safetensors",
|
| 903 |
+
"semantic_tokenizer_encoder.stem.stage.2.ffn_norm.weight": "model-00007-of-00008.safetensors",
|
| 904 |
+
"semantic_tokenizer_encoder.stem.stage.2.gamma": "model-00007-of-00008.safetensors",
|
| 905 |
+
"semantic_tokenizer_encoder.stem.stage.2.mixer.conv.bias": "model-00007-of-00008.safetensors",
|
| 906 |
+
"semantic_tokenizer_encoder.stem.stage.2.mixer.conv.weight": "model-00007-of-00008.safetensors",
|
| 907 |
+
"semantic_tokenizer_encoder.stem.stage.2.norm.weight": "model-00007-of-00008.safetensors"
|
| 908 |
+
}
|
| 909 |
+
}
|
processor_config.json
ADDED
|
@@ -0,0 +1,18 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"audio_bos_token": "<|object_ref_start|>",
|
| 3 |
+
"audio_duration_token": "<|AUDIO_DURATION|>",
|
| 4 |
+
"audio_eos_token": "<|object_ref_end|>",
|
| 5 |
+
"audio_token": "<|box_start|>",
|
| 6 |
+
"feature_extractor": {
|
| 7 |
+
"eps": 1e-06,
|
| 8 |
+
"feature_extractor_type": "VibeVoiceAcousticTokenizerFeatureExtractor",
|
| 9 |
+
"feature_size": 1,
|
| 10 |
+
"normalize_audio": true,
|
| 11 |
+
"padding_side": "right",
|
| 12 |
+
"padding_value": 0.0,
|
| 13 |
+
"return_attention_mask": true,
|
| 14 |
+
"sampling_rate": 24000,
|
| 15 |
+
"target_dB_FS": -25
|
| 16 |
+
},
|
| 17 |
+
"processor_class": "VibeVoiceAsrProcessor"
|
| 18 |
+
}
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3fd169731d2cbde95e10bf356d66d5997fd885dd8dbb6fb4684da3f23b2585d8
|
| 3 |
+
size 11421892
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"bos_token": null,
|
| 5 |
+
"clean_up_tokenization_spaces": false,
|
| 6 |
+
"eos_token": "<|endoftext|>",
|
| 7 |
+
"errors": "replace",
|
| 8 |
+
"extra_special_tokens": [
|
| 9 |
+
"<|im_start|>",
|
| 10 |
+
"<|im_end|>",
|
| 11 |
+
"<|object_ref_start|>",
|
| 12 |
+
"<|object_ref_end|>",
|
| 13 |
+
"<|box_start|>",
|
| 14 |
+
"<|box_end|>",
|
| 15 |
+
"<|quad_start|>",
|
| 16 |
+
"<|quad_end|>",
|
| 17 |
+
"<|vision_start|>",
|
| 18 |
+
"<|vision_end|>",
|
| 19 |
+
"<|vision_pad|>",
|
| 20 |
+
"<|image_pad|>",
|
| 21 |
+
"<|video_pad|>"
|
| 22 |
+
],
|
| 23 |
+
"is_local": false,
|
| 24 |
+
"model_max_length": 131072,
|
| 25 |
+
"pad_token": "<|endoftext|>",
|
| 26 |
+
"processor_class": "VibeVoiceAsrProcessor",
|
| 27 |
+
"split_special_tokens": false,
|
| 28 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 29 |
+
"unk_token": null
|
| 30 |
+
}
|