Back: Catalog

speaches

Speeches is a OpenAI API compatible server software for audio production and speech-to-text conversion operations. It operates on the basis of faster-whisper, piper, and Kokoro models; it supports fluid transcription and dynamic model management features. It can be distributed with Docker and run on CPU and GPU hardware.

SP

speaches

New80

speaches-ai · Containers & Kubernetes

STARS3.7K
FORKS464
COMMITS0
CONTRIBUTORS0
VERSIONv0.9.0-rc.3
UPDATEDyesterday
PythonOpenAI APIText-to-SpeechSpeech-to-TextMIT

Common commercial products in this category: Docker Desktop, Red Hat OpenShift, Amazon EKS

What can it do?
  • You can convert text to audio through OpenAI API compatible clients and SDK.
  • Using the Faster-whisper infrastructure, you can perform fluid audio transcription through the server.
  • With Piper and Kokoro models, you can manage text-to-sound and dynamic model loading.
User reviews
  • • NeutralGitHub Issue

    Write an Open WebUI integration tutorial

    “If someone wants to take this on please let me know”

  • • NeutralGitHub Issue

    PRELOAD_MODELS doesn't work on last Docker image tag (but with build locally, it works)

    “With local deployment, the PRELOAD_MODELS config variable works perfectly : ``` PRELOAD_MODELS='["Systran/faster-whisper-medium.en", "Systran/faster-whisper-small.en"]' MAX_MODELS=2 uvicorn main:app --port 8080 --log-level debug --reload ``` ![image]( But in a docker compose…”

Website GitHub

My note

Alternative pages

Docker Desktop alternativesRed Hat OpenShift alternativesAmazon EKS alternatives

More in: Containers & Kubernetes