speaches
CLAUDE.md3.7k starsChanged 7 months agoMITView on GitHub
Trending
Steady talk.about the whole repository
Who's talking about speaches-ai/speaches
- zugaldia/speedofsound“Streaming mode (a.k.a. real-time mode, live mode)”GitHub · May 30, 2026 · 4 reactions
- DevEmperor/DictateKeyboard“External Server Integration Fails – “Custom Server” Feature Not Working”GitHub · Jun 19, 2025 · 3 reactions
- Demizo/Daily_You“[Feature Request] Audio Logs”GitHub · Mar 7, 2025 · 2 reactions
- modelscope/FunASR“Track speaches FunASR review wait”GitHub · Jun 30, 2026 · 1 reactions
- beclab/apps“[Add App] speaches”GitHub · Jan 23, 2026 · 1 reactions
The biggest 5 of 27 mentions linking to the repository that our crawl found. Each line is their own headline.
See all 27
Talked about on GitHub (27 times).
On GitHub · 27
- May 30, 2026Streaming mode (a.k.a. real-time mode, live mode) — zugaldia/speedofsound4 reactions · 4 comments
- Jun 19, 2025External Server Integration Fails – “Custom Server” Feature Not Working — DevEmperor/DictateKeyboard3 reactions · 14 comments
- Mar 7, 2025[Feature Request] Audio Logs — Demizo/Daily_You2 reactions · 4 comments
- Jun 30, 2026Track speaches FunASR review wait — modelscope/FunASR1 reaction
- Jan 23, 2026[Add App] speaches — beclab/apps1 reaction · 1 comment
22 more on GitHub
- Aug 3, 2025Automatic Speech Recognition (ASR) — hongbo-miao/hongbomiao.com1 reaction
- Nov 6, 2026feat: Speaches 语音识别与合成插件(CUDA 12.8 自建镜像) — geraldpeng6/wonton-plugin-speaches0 reactions
- Sep 13, 2026Speaches — Evaluar mejora de voz en español frente a Hermes nativo — bobymor1-dev/hermes-component-radar0 reactions
- Sep 12, 2026[deferred] voxora-openai-asr: generic OpenAI-compat engine for OpenAI / Groq / Together / Mistral / local-server cluster — airvzxf/voxora0 reactions · 2 comments
- Aug 31, 2026feat: add Speaches CUDA speech-to-text stack — papatenko/homelab0 reactions
- Jun 8, 2026DGX Speaches Docker deploy — install hardening + pre-warm + image pinning — chipi/podcast_scraper0 reactions · 1 comment
- May 4, 2026[Issue]: Broken TTS since 1.5.7 — Pasta-Devs/Marinara-Engine0 reactions
- Apr 1, 2026Add project: speaches — mahmoud/zerover0 reactions
- Jan 10, 2026feat: Explore speaker diarization for multi-speaker real-time translation — kizuna-ai-lab/sokuji-tsuyaku-api0 reactions
- Jan 5, 2026Implement a realtime audio caption and translation — translate-tools/linguist0 reactions · 4 comments
- Oct 10, 2025Contribute to Speaches issue #547 for Voxtral integration — OlivierAlbertini/Voxtral-WebUI0 reactions
- Aug 25, 2025Text to speech (TTS) and Speech to text (STT) API — TU-Wien-dataLAB/aqueduct0 reactions
- Jul 21, 2025STT/Whisper: "language" parameter should not be sent — Kochava-Studios/witsy0 reactions · 6 comments
- Jul 17, 2025Add OpenAI TTS support — steveseguin/social_stream0 reactions · 8 comments
- Jun 3, 2025ghcr.io/speaches-ai/speaches:latest-cpu — DaoCloud/public-image-mirror0 reactions · 1 comment
- May 28, 2025Handle fragmentation of WebRTC datachannel messages to avoid running into browser limits — speaches-ai/realtime-console0 reactions
- Mar 26, 2025Use speaches to test generateSpeech / generateTranscription — moeru-ai/xsai0 reactions
- Mar 20, 2025Update README.md with speaches instead of faster-whisper-server — SYSTRAN/faster-whisper0 reactions
- Feb 20, 2025Check out faster whisper server — shuffle-project/melvin-asr0 reactions · 1 comment
- Jan 26, 2025Incorrect dependency constraint: `onnxruntime-gpu` isn't available on ARM — thewh1teagle/kokoro-onnx0 reactions
- Jan 17, 2025Add OpenAI-compatible server configuration for Whisper — thewh1teagle/vibe0 reactions · 3 comments
- Jan 16, 2025Add support for OpenAI compatable APIs — meain/shuh0 reactions
The file
Show its rules and text
We split this CLAUDE.md into the rules it gives an agent
14 rules. About 0 are common, 5 are distinctive, and 9 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈5
Distinctive · estimate
This author's own choices. Worth a look.
9
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 1
Always: use type hints for function/method parameters and return types.
- Distinctive: line 2
Use latest type hinting format (Python 3.12).
- Very specific: line 3
Use list[str] instead of List[str].
- Very specific: line 7
Prefer using pathlib module over os.path for file and path manipulations.
- Very specific: line 8
Prefer using pydantic.BaseModel over dataclasses.dataclass for data validation and serialization.
- Very specific: line 9
Use logger.exception for logging exceptions with stack traces instead of logger.error(f"Error occured: {e}")
- Very specific: line 10
Use logger.xxx() for logging instead of print()
- Distinctive: line 11
Do not use emojis or any special characters in code comments or log messages.
- Very specific: line 12
Use f-strings for string formatting instead of str.format() or concatenation.
- Distinctive: line 13
Do not write docsctrings
- Distinctive: line 14
Prefer defining functions over classes with methods when state is not needed.
- Very specific: line 15
Use pydantic_settings.BaseSettings for configuration settings.
- Very specific: line 16
Always run ruff format <modified files> and ruff check <modified files> (add --fix to ruff check to auto-fix some of the issues).
- Very specific: line 17
Use semantic git commit messages (e.g., feat: add new feature, fix: correct a bug, docs: update documentation).
≈ “Common” is an estimate: at least 1 in 5 of the 27,641 CLAUDE.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
- Always use type hints for function/method parameters and return types.
- Use latest type hinting format (Python 3.12). For example:
- Use
list[str]instead ofList[str]. - Use
dict[str, int]instead ofDict[str, int]. - Use
tuple[int, ...]instead ofTuple[int, ...]. - Use
set[str]instead ofSet[str].
- Use
- Prefer using
pathlibmodule overos.pathfor file and path manipulations. - Prefer using
pydantic.BaseModeloverdataclasses.dataclassfor data validation and serialization. - Use
logger.exceptionfor logging exceptions with stack traces instead oflogger.error(f"Error occured: {e}") - Use
logger.xxx()for logging instead ofprint() - Do not use emojis or any special characters in code comments or log messages.
- Use f-strings for string formatting instead of
str.format()or concatenation. - Do not write docsctrings
- Prefer defining functions over classes with methods when state is not needed.
- Use
pydantic_settings.BaseSettingsfor configuration settings. - Always run
ruff format <modified files>andruff check <modified files>(add--fixtoruff checkto auto-fix some of the issues). - Use semantic git commit messages (e.g.,
feat: add new feature,fix: correct a bug,docs: update documentation).
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

