Audio
Whisper
Robust Speech Recognition via Large-Scale Weak Supervision
★ 109.9k starsMIT
View guide →// audio · PaddlePaddle/PaddleSpeech
PaddleSpeech
PaddleSpeech is an open-source audio project by PaddlePaddle (PaddlePaddle/PaddleSpeech on GitHub). It has 12,690 GitHub stars and 1,955 forks, is released under the Apache-2.0 license, and was last updated on GitHub on 12 August 2026.
Easy-to-use Speech Toolkit including Self-Supervised Learning model, SOTA/Streaming ASR with punctuation, Streaming TTS with text…
★ 12.7k stars⑂ 2.0k forksApache-2.0updated 2026-08
▦ visual — this project's README ships no embeddable screenshot,
so here's a placeholder. The guide below is complete regardless.
so here's a placeholder. The guide below is complete regardless.
// project facts
Project facts
- Repository
- PaddlePaddle/PaddleSpeech
- GitHub stars
- 12,690
- Forks
- 1,955
- License
- Apache-2.0
- Last updated on GitHub
- 12 August 2026
- Category
- Audio
- Listing
- Full install guide
- Official site
- Official site
// how it works
How it works
- Pick your direction
Speech-to-text, text-to-speech, voice cloning, or music/SFX generation — each project here has a lane.
- Install the runtime
Python packages, Docker images, and C++ builds cover servers, desktops, and edge devices.
- Load a model or voice
Download the weights or pick a voice; many tools bundle a strong default so you can start immediately.
- Process your audio
Transcribe files, synthesize scripts, or generate music — then export standard audio formats.
// features
Why people love it
State-of-the-art transcription
Open speech recognition now rivals paid APIs on accuracy across languages.
Natural neural voices
Modern TTS is strikingly human — good enough for narration and voiceovers.
Voice cloning
Clone a voice from a short sample for consistent characters and branding.
Music & SFX generation
Compose backing tracks and sound effects from text descriptions.
Runs offline
Entire pipelines work without internet — ideal for private or field use.
API-compatible servers
Drop-in endpoints that mimic commercial APIs, so existing apps switch easily.
Advertisement
// run it
Run it yourself
- Step 1 — Clone the repo
Download the source and move into the project folder.
git clone https://github.com/PaddlePaddle/PaddleSpeech.git cd PaddleSpeech - Step 2 — Create an environment
Isolate dependencies so nothing clashes with your other projects.
python -m venv .venv source .venv/bin/activate # Windows: .venv\Scripts\activate - Step 3 — Install and launch
Install the requirements, then follow the README's launch command (it varies per project).
pip install -r requirements.txt
// use cases
What you can build
- Transcribe meetings, interviews, and podcasts.
- Narrate audiobooks and video voiceovers.
- Build voice interfaces for apps and robots.
- Generate royalty-free music and effects.
Advertisement
// faq
Questions, answered
How accurate is open-source transcription?
Excellent — the leading open models match commercial services on clean audio, in many languages.
Can I clone a voice ethically?
The tech clones from short samples; only ever clone voices you own or have explicit permission to use.
Who maintains PaddleSpeech?
It is an open-source project at PaddlePaddle/PaddleSpeech with 1,955 forks. Check the repo's contributors and recent commits to see how active maintenance is.
Do I need a GPU?
Transcription and TTS run on CPU acceptably; GPUs speed up batch jobs and music generation a lot.
Which languages are supported?
Coverage varies — transcription is broadest, TTS quality is best in major languages.
Can I use outputs commercially?
Usually yes, but check both the code license and the model license.
// keep exploring
Similar tools
Audio
whisper.cpp
Port of OpenAI's Whisper model in C/C++
★ 54.1k starsMIT
View guide →Audio
Faster Whisper
Faster Whisper transcription with CTranslate2
★ 25.7k starsMIT
View guide →Audio
Coqui TTS
Deep learning toolkit for text-to-speech, battle-tested in research and production
★ 46.1k starsMPL-2.0
View guide →Audio
Fish Speech
Fish Speech — Fish Audio's expressive open text-to-speech and voice cloning: SOTA-quality TTS you can self-host, under the Fish Audio Community License (check terms)
★ 33.3k starsCustom (check repo)
View guide →Audio
vits
VITS — the original end-to-end text-to-speech model: a conditional VAE with adversarial learning, in PyTorch.
★ 7.9k starsMIT
View guide →// alternatives
Alternatives in Audio
The most-starred other Audio projects in the LibreAI catalogue — the realistic shortlist if PaddleSpeech is not the right fit.
- MoneyPrinterTurbo★ 128.3k
利用 AI 大模型和自动化工作流，根据主题或关键词一键生成高清短视频。Generate HD short videos from a topic or keyword with an automated AI workflow.
- Whisper★ 109.9k
Robust Speech Recognition via Large-Scale Weak Supervision
- GPT-SoVITS★ 62.3k
1 min voice data can also be used to train a good TTS model! (few shot voice cloning)
- Real-Time-Voice-Cloning★ 60.2k
Clone a voice from 5 seconds of audio and generate arbitrary speech in real time
- whisper.cpp★ 54.1k
Port of OpenAI's Whisper model in C/C++
// more in audio
More in Audio
Nearby Audio projects in the catalogue, closest in stars to PaddleSpeech — keep exploring the shelf either side of it.
- voice-pro★ 13.0k
Gradio WebUI for creators and developers, featuring key TTS (Edge-TTS, kokoro) and zero-shot Voice Cloning (E2 & F5-TTS, CosyVoice), with Whisper audio proce…
- edge-tts★ 12.2k
Use Microsoft Edge's online text-to-speech service from Python WITHOUT needing Microsoft Edge or Windows or an API key
- Qwen3-TTS★ 13.6k
Alibaba's open voice models: expressive, streaming-capable speech with free-form voice design.
- speechbrain★ 11.9k
A PyTorch-based Speech Toolkit
- sherpa-onnx★ 15.1k
Speech-to-text, text-to-speech, speaker diarization, speech enhancement, source separation, and VAD using next-gen Kaldi with onnxruntime without Internet co…
- Moshi★ 11.4k
Moshi — Kyutai full-duplex spoken dialogue model: talk and listen at the same time, ~200ms latency, Apache-2.0
Get PaddleSpeech
Free and open — grab the source, star the repo, and start creating.