Taproot

Latest version: v0.3.8

Safety actively analyzes 723158 Python packages for vulnerabilities to keep your Python projects secure.

Scan your dependencies

Page 2 of 3

0.3.2

Added Zonos Text-to-Speech, in Hybrid and Transformer versions.

<h3>zonos-hybrid</h3>
<table><tbody><tr><td>Name</td><td>ZonosHybridSpeechSynthesis</td></tr><tr><td>Author</td><td>Zyphra Team<br /><a href="https://www.zyphra.com/post/beta-release-of-zonos-v0-1" target="_blank">https://www.zyphra.com/post/beta-release-of-zonos-v0-1</a></td></tr><tr><td>License</td><td><a href="https://www.apache.org/licenses/LICENSE-2.0" target="_blank">Apache License 2.0</a></td></tr><tr><td>Files</td><td><ol><li><a href="https://huggingface.co/benjamin-paine/taproot-common/resolve/main/speech-synthesis-zonos-hybrid-v0-1.bf16.safetensors" target="_blank">speech-synthesis-zonos-hybrid-v0-1.bf16.safetensors (3.30 GB)</a></li><li><a href="https://huggingface.co/benjamin-paine/taproot-common/resolve/main/audio-vocoder-descript-44khz.safetensors" target="_blank">audio-vocoder-descript-44khz.safetensors (306.51 MB)</a></li><li><a href="https://huggingface.co/benjamin-paine/taproot-common/resolve/main/audio-diarisation-zonos-speaker-embedding.safetensors" target="_blank">audio-diarisation-zonos-speaker-embedding.safetensors (396.35 MB)</a></li></ol><p><strong>Total Size</strong>: 4.01 GB</p></td></tr><tr><td>Minimum VRAM</td><td>4.04 GB</td></tr></tbody></table>
<h3>zonos-transformer</h3>
<table><tbody><tr><td>Name</td><td>ZonosTransformerSpeechSynthesis</td></tr><tr><td>Author</td><td>Zyphra Team<br /><a href="https://www.zyphra.com/post/beta-release-of-zonos-v0-1" target="_blank">https://www.zyphra.com/post/beta-release-of-zonos-v0-1</a></td></tr><tr><td>License</td><td><a href="https://www.apache.org/licenses/LICENSE-2.0" target="_blank">Apache License 2.0</a></td></tr><tr><td>Files</td><td><ol><li><a href="https://huggingface.co/benjamin-paine/taproot-common/resolve/main/speech-synthesis-zonos-transformer-v0-1.bf16.safetensors" target="_blank">speech-synthesis-zonos-transformer-v0-1.bf16.safetensors (3.25 GB)</a></li><li><a href="https://huggingface.co/benjamin-paine/taproot-common/resolve/main/audio-vocoder-descript-44khz.safetensors" target="_blank">audio-vocoder-descript-44khz.safetensors (306.51 MB)</a></li><li><a href="https://huggingface.co/benjamin-paine/taproot-common/resolve/main/audio-diarisation-zonos-speaker-embedding.safetensors" target="_blank">audio-diarisation-zonos-speaker-embedding.safetensors (396.35 MB)</a></li></ol><p><strong>Total Size</strong>: 3.95 GB</p></td></tr><tr><td>Minimum VRAM</td><td>4.04 GB</td></tr></tbody></table>

0.3.1

Added Deepseek R1 (Llama 8B Distillation)

Installation

This is for the Q8 quantization. Omit `q8-0` for FP16, or use `q6-k`, `q5-k-m`, etc.


taproot install text-generation:deepseek-r1-llama-8b-q8-0


Chat (Commandline)


taproot chat deepseek-r1-llama-8b-q8-0

[system] Model set to deepseek-r1-llama-8b-q8-0.
[system] Seed set to 1296776529.
[user] How many R's are in the word strawberry? Reason out your answer step-by-step.
[assistant] <think>
Okay, so I need to figure out how many R's are in the word "strawberry." Let me start by writing down the word and breaking it down letter by letter. S-T-R-A-W-B-E-R-R-Y. Hmm, that doesn't look right. Wait, maybe I should write it correctly first.

Strawberry is spelled S-T-R-A-W-B-E-R-R-Y. Now, let's go through each letter one by one to count the R's. Starting from the beginning: S, T, R... so that's one R. Then A, W, B, E... no more R's there. Next comes another R after E, so that's two R's now. And then another R at the end before the Y. So that would make it three R's in total.

Wait, did I count correctly? Let me check again. S(1), T(2), R(3) – first R. Then A(4), W(5), B(6), E(7). Next is R(8), so that's the second R. And then another R at the end before Y, which would be the third R. So yes, three R's.

I think I got it right. But just to make sure, maybe I can visualize the word or write it out again. S T R A W B E R R Y. Yep, there are three R's: one after T, one after E, and another before Y.
</think>

The word "strawberry" has three R's.

**Step-by-Step Explanation:**

1. **Write Down the Word:** Start by writing the word to visualize each letter clearly.
- S T R A W B E R R Y

2. **Identify Each Letter Position:**
- 1: S
- 2: T
- 3: R (First R)
- 4: A
- 5: W
- 6: B
- 7: E
- 8: R (Second R)
- 9: R (Third R)
- 10: Y

3. **Count the R's:**
- The third letter is R.
- The eighth letter is R.
- The ninth letter is R.

4. **Total Count:** There are three R's in total.

**Answer:** There are 3 R's in the word "strawberry.

0.3.0

1. Added `http` and `https` protocol, used the same as `ws` with better transfer rates for large files.
2. Made default protocol TCP.
3. Made websockets optional.
4. Changed how executors are spawned. This should improve overall stability, and promote cleaner shutdowns.

To install dependencies for http:

`pip install taproot[http]`

To install dependencies for websockets:

`pip install taproot[ws]`

0.2.3

- Further improvements to latency and bandwidth in TCP and Unix configurations.
- Added speed test and published results.

0.2.0

- Removed dependence on re-entrant loops (dropped `nest_asyncio` requirement)
- Made `uvloop` the default loop engine, when installed
- Added `packages` and `files` commands for clutter-free requirement file generation (see below)

sh
taproot packages speech-synthesis:kokoro text-generation:llama-v3-1-8b-instruct-q6-k


accelerate~=1.0
torchvision<0.20,>=0.19
Pillow~=9.5
torch<2.5,>=2.4
numpy~=1.22
torchaudio<2.5,>=2.4
librosa>=0.10
einops>=0.8
scipy==1.11.2
safetensors~=0.4
transformers>=4.41
scikit-learn~=1.5
phonemizer~=3.3
llama-cpp-python>=0.2.79

*Note: some packages may not be installable via pip, like `dwpose`, which requires `mim`. The recommended installation method is still to use `taproot install`, which takes care of this edge case and many more.*

sh
taproot files speech-synthesis:kokoro text-generation:llama-v3-1-8b-instruct-q6-k


https://huggingface.co/benjamin-paine/taproot-common/resolve/main/text-generation-llama-v3-1-8b-instruct-q6-k.gguf
https://huggingface.co/benjamin-paine/taproot-common/resolve/main/speech-synthesis-kokoro-v0-19.safetensors
https://huggingface.co/benjamin-paine/taproot-common/resolve/main/speech-synthesis-kokoro-v0-19-voices.safetensors

0.1.4

Fixed promise handling for some error cases.

Page 2 of 3

© 2025 Safety CLI Cybersecurity Inc. All Rights Reserved.