经 AI Skill Hub 精选评估,whisper-flamingo — AI 语音识别工具中文文档 获评「推荐使用」。这款AI工具在功能完整性、社区活跃度和易用性方面表现出色,AI 评分 7.8 分,适合有一定技术背景的用户使用。
whisper-flamingo — AI 语音识别工具中文文档 是一款基于 Jupyter Notebook 开发的开源工具,专注于 stt 等核心功能。作为 GitHub 开源项目,它拥有活跃的社区支持和持续的版本迭代,代码完全透明可审计,支持本地部署以保护数据隐私。无论是个人使用还是集成到企业工作流,都能提供稳定可靠的解决方案。
whisper-flamingo — AI 语音识别工具中文文档 是一款基于 Jupyter Notebook 开发的开源工具,专注于 stt 等核心功能。作为 GitHub 开源项目,它拥有活跃的社区支持和持续的版本迭代,代码完全透明可审计,支持本地部署以保护数据隐私。无论是个人使用还是集成到企业工作流,都能提供稳定可靠的解决方案。
# 克隆仓库 git clone https://github.com/roudimit/whisper-flamingo cd whisper-flamingo # 查看安装说明 cat README.md # 按 README 完成环境依赖安装后即可使用
# 查看帮助 whisper-flamingo --help # 基本运行 whisper-flamingo [options] <input> # 详细使用说明请查阅文档 # https://github.com/roudimit/whisper-flamingo
# whisper-flamingo 配置说明 # 查看配置选项 whisper-flamingo --config-example > config.yml # 常见配置项 # output_dir: ./output # log_level: info # workers: 4 # 环境变量(覆盖配置文件) export WHISPER_FLAMINGO_CONFIG="/path/to/config.yml"
We propose Whisper-Flamingo which integrates visual features into the Whisper speech recognition and translation model with gated cross attention. Our audio-visual Whisper-Flamingo outperforms audio-only Whisper on English speech recognition and En-X translation for 6 languages in noisy conditions. Moreover, Whisper-Flamingo is a versatile model and conducts all of these tasks using one set of parameters, while prior methods are trained separately on each language.

mWhisper-Flamingo for Multilingual Audio-Visual Noise-Robust Speech Recognition
We propose mWhisper-Flamingo for multilingual AVSR. To enable better multi-modal integration and improve the noisy multilingual performance, we introduce decoder modality dropout where the model is trained both on paired audio-visual inputs and separate audio/visual inputs. mWhisper-Flamingo achieves state-of-the-art WER on MuAViC, an AVSR dataset of 9 languages.

pip install -r requirements.txt
python -m pip install pip==24.0 pip --version git clone -b muavic https://github.com/facebookresearch/av_hubert.git cd av_hubert git submodule init git submodule update
cd fairseq pip install --editable ./ cd ../..
Install extra packages used in our project: pip install tiktoken==0.5.2 pytorch-lightning==2.1.3 numba==0.58.1 transformers==4.36.2 evaluate tensorboardX ```
Important: to use mWhisper-Flamingo, a minor change is required in the AV-HuBERT code. Specfically, comment out line 624 and add this after line 625: features_audio = torch.zeros_like(features_video). This is needed since we only use video inputs with AV-HuBERT, not audio. Otherwise you will get an error about 'NoneType' object.
Check out the video demo below (turn sound on). We made several videos about Whisper-Flamingo: - 30s demo of Whisper-Flamingo (same video below): YouTube link - 2m demo comparing Whisper and Whisper-Flamingo: YouTube link - 10m presentation: YouTube link
- 1m demo of mWhisper-Flamingo (turn sound on). YouTube link. - 20m presentation: YouTube link. <table class="center"> <tr> <td width=100% style="border: none"> <video controls autoplay loop src="https://github.com/user-attachments/assets/4852942f-e54d-4bf3-9f28-e038b4edb100" muted="false"></video> </td> </tr> </table>
Whisper-Flamingo Colab notebooks. Note: Google Colab updated Python to 3.11, which makes the Fairseq version incompatible and the notebooks might not work anymore. We recommend installing our conda environment locally and running the local versions. - Test Whisper-Flamingo on an example audio / video . Local copy:
notebooks/whisper_flamingo_demo_noise.ipynb. - Reproduce our results on LRS3 / MuAViC: . Local copy:
notebooks/whisper_flamingo_demo.ipynb.
mWhisper-Flamingo notebooks: - Test mWhisper-Flamingo on an example audio / video (reproduce the demo video): notebooks/mwhisper_flamingo_demo_noise.ipynb. - Reproduce our results on MuAViC Es ASR: notebooks/mwhisper_flamingo_demo.ipynb
Since this project uses the MuAViC dataset, we base our virtual environment on theirs.
Create a fresh virtual environment:
conda create -n whisper-flamingo python=3.8 -y
conda activate whisper-flamingo Clone MuAViC repo and install their requirements: conda install -c conda-forge ffmpeg==4.2.2 -y
conda install -c conda-forge sox -y
git clone https://github.com/facebookresearch/muavic.git muavic-setup
cd muavic-setup
pip install -r requirements.txt
cd .. Clone the "muavic" branch of av_hubert's repo and install Fairseq: ```
该工具使用 NOASSERTION 协议,商用场景请仔细阅读协议条款,必要时咨询法律意见。
AI Skill Hub 为第三方内容聚合平台,本页面信息基于公开数据整理,不对工具功能和质量作任何法律背书。
建议在沙箱或测试环境中充分验证后,再部署至生产环境,并做好必要的安全评估。
📄 NOASSERTION — 请查阅原始协议条款了解具体使用限制。
AI Skill Hub 点评:whisper-flamingo — AI 语音识别工具中文文档 的核心功能完整,质量良好。对于AI 技术爱好者来说,这是一个值得纳入个人工具库的选择。建议先在非生产环境试用,再逐步推广。
| 原始名称 | whisper-flamingo |
| 原始描述 | Whisper-Flamingo [Interspeech 2024] and mWhisper-Flamingo [IEEE SPL 2025] for Audio-Visual Speech Recognition and Translation |
| Topics | stt |
| GitHub | https://github.com/roudimit/whisper-flamingo |
| License | NOASSERTION |
| 语言 | Jupyter Notebook |
收录时间:2026-05-22 · 更新时间:2026-05-22 · License:NOASSERTION · AI Skill Hub 不对第三方内容的准确性作法律背书。