Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[WIP] Fetch upstream #6

Draft
wants to merge 167 commits into
base: tokenize_hotwords_flag_master
Choose a base branch
from

Conversation

w11wo
Copy link

@w11wo w11wo commented Dec 12, 2024

No description provided.

csukuangfj and others added 30 commits October 9, 2024 12:01
If the input audio file is less than 10 seconds long, there is only 
one chunk, and there is no need to compute embeddings or 
do clustering.

We can use the segmentation result from the speaker segmentation 
model directly.
#1408 uses [node-addon-api](https://github.com/nodejs/node-addon-api) to call C API from JavaScript, whereas this pull request uses WebAssembly to call C API from JavaScript.
… be found for specific aarch64 cross-compilation environments (#1436)

Co-authored-by: xiao <shawl336@163.com>
---------

Co-authored-by: root <1552138571@qq.com>
It is a wrapper around the C API.
csukuangfj and others added 30 commits January 3, 2025 17:09
…1699)

由于SenseVoiceSmall.from_pretrained()
    调用的funasr.auto.auto_model.AutoModel.build_model()默认device是cuda
    (在cuda available的环境中)
    ```py
    device = kwargs.get("device", "cuda")
    if not torch.cuda.is_available() or kwargs.get("ngpu", 1) == 0:
        device = "cpu"
        kwargs["batch_size"] = 1
    kwargs["device"] = device
    ```
    而export-onnx.py里的tensor默认都是cpu, 导致
    RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda:0 and cpu
    所以直接在加载model的时候指定cpu
See also
#1695 (comment)

We need to place a 0 at the end of the buffer.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.