项目已并入 Flashlight，后续开发在 Flashlight 的 ASR 应用中；复现结果必须使用 Flashlight <= 0.3.2，构建 recipes 需用 Flashlight 0.3 分支并带 ASR 应用；Flashlight 或 ArrayFire 装在非标准路径时需额外传 -Dflashlight_DIR 与 -DArrayFire_DIR
wav2letter
flashlight活跃维护Facebook AI 开源的端到端语音识别训练工具包，用 C++ 实现，可基于 wav2letter 架构训练声学模型，省去传统 ASR 繁琐的特征工程流程。
Facebook AI Research's Automatic Speech Recognition Toolkit
v0.2发布于 2020-12-28 17:17 UTC6,445star982fork同步于 2026-11-10 17:35 UTC
mkdir build && cd build
cmake .. && make -j8 硬件要求按 README 说明
README 未说明显存、CPU 或 GPU 型号要求
Facebook 开源的端到端语音识别工具包，提供可复现论文的 recipes 与预训练模型
Facebook AI 开源的端到端语音识别训练工具包，用 C++ 实现，可基于 wav2letter 架构训练声学模型，省去传统 ASR 繁琐的特征工程流程。
核心能力
- 提供多篇 ASR 论文的可复现 recipes 与预训练模型
- 覆盖流式卷积、半监督、自训练、无词典等方向
- 训练与评测的数据准备脚本集中在 data 目录
- 基于 Flashlight 与 ArrayFire 的 C++ 实现
容易踩的坑
内容由模型根据 README 生成，可能不完整或不准确，以官方文档为准。看原文 README
- v0.2v0.2 (pre Flashlight-consolidation)最新正式版2020-12-28 · 2020-12-28 17:17 UTC
This is an old release and is the final release before the wav2letter was consolidated into Flashlight. See the ASR application in Flashlight for the new wav2letter source.
- v0.1Initial Release2018-12-21 · 2018-12-21 21:58 UTC
查看更新说明
See the docs for more details.
Thanks to our contributors:
Vineel Pratap, Awni Hannun, Qiantong Xu, Jacob Kahn, Jeff Cai, Gabriel Synnaeve, Vitaliy Liptchinsky, Ronan Collobert, Ann Lee, Tatiana Likhomanenko and Jay Mahadeokar
用户点评
登录用户的真实使用记录，可按版本和硬件筛选共 0 条
还没有人点评。第一个说出真实体验的人最有价值。
wav2letter++
Important Note:
wav2letter has been moved and consolidated into Flashlight in the ASR application.
Future wav2letter development will occur in Flashlight.
To build the old, pre-consolidation version of wav2letter, checkout the wav2letter v0.2 release, which depends on the old Flashlight v0.2 release. The wav2letter-lua project can be found on the wav2letter-lua branch, accordingly.
For more information on wav2letter++, see or cite this arXiv paper.
Recipes
This repository includes recipes to reproduce the following research papers as well as pre-trained models. All results reproduction must use Flashlight <= 0.3.2 for exact reproducability. Papers contained here include:
- Pratap et al. (2020): Scaling Online Speech Recognition Using ConvNets
- Synnaeve et al. (2020): End-to-end ASR: from Supervised to Semi-Supervised Learning with Modern Architectures
- Kahn et al. (2020): Self-Training for End-to-End Speech Recognition
- Likhomanenko et al. (2019): Who Needs Words? Lexicon-free Speech Recognition
- Hannun et al. (2019): Sequence-to-Sequence Speech Recognition with Time-Depth Separable Convolutions
Data preparation for training and evaluation can be found in data directory.
Building the Recipes
First, install Flashlight (using the 0.3 branch is required) with the ASR application.
mkdir build && cd build
cmake .. && make -j8
If Flashlight or ArrayFire are installed in nonstandard paths via a custom CMAKE_INSTALL_PREFIX, they can be found by passing
-Dflashlight_DIR=[PREFIX]/usr/share/flashlight/cmake/ -DArrayFire_DIR=[PREFIX]/usr/share/ArrayFire/cmake
when running cmake.
Join the wav2letter community
- Facebook page: https://www.facebook.com/groups/717232008481207/
- Google group: https://groups.google.com/forum/#!forum/wav2letter-users
- Contact: [email protected], [email protected], [email protected], [email protected], [email protected], [email protected], [email protected], [email protected], [email protected]
License
wav2letter++ is MIT-licensed, as found in the LICENSE file.
VoiceStudio
debpalash·Python
本地运行的 ElevenLabs 开源替代品，支持声音克隆、语音设计、视频配音、听写、转写和有声书制作，覆盖 646 种语言。
语音克隆+4语音推理
107004+388012kv0.5.72026-11-10 16:38 UTC
voicebox
jamiepine·Python
开源 AI 语音工作室，支持声音克隆、语音听写和语音生成，基于 Qwen3-TTS 与 Whisper，可在 CUDA 或 MLX 上本地运行。
语音克隆+4语音推理
60376+6237.6kv0.5.02026-10-07 01:17 UTC
whisper.cpp
ggml-org·C++
OpenAI Whisper 语音识别模型的 C/C++ 移植版，无需 Python 和 GPU 依赖，在普通 CPU 上就能跑语音转文字。
语音识别+4语音推理
55071+846.3kv1.9.52026-11-10 12:09 UTC
faster-whisper
SYSTRAN·Python
基于 CTranslate2 重实现的 Whisper 语音转文字推理引擎，通过量化与高效推理大幅提速并降低显存占用，适合本地部署转录。
语音识别+4语音推理
26650+782.2kv1.2.12026-11-10 10:13 UTC