K07VN/capcut-tts-api

What it solves

This project provides a Python interface to interact with CapCut's speech services. It simplifies tasks like converting text to speech, transcribing audio or video files into text, and translating speech, allowing developers to automate these workflows programmatically rather than manually.

How it works

The library acts as a client that communicates with CapCut's speech APIs. It handles the full lifecycle of a task: submitting requests, polling for completion, and downloading the resulting audio or subtitle files. It includes helper methods for common operations like generating audio from SRT subtitle files or performing a round-trip of generating speech and then transcribing it back to text.

Who it’s for

It is designed for Python developers who want to integrate CapCut's speech-to-text and text-to-speech capabilities into their own applications, scripts, or automation pipelines.

Highlights

• Text-to-Speech (TTS): Create audio from text or SRT files with support for various voices.
• Speech-to-Text (STT): Transcribe local audio or video files and export results to SRT format.
• Translation: Request translations during the transcription process.
• Flexible Control: Use high-level convenience methods for simple tasks or low-level calls for managing specific task IDs and polling manually.
• Command-Line Interface: Includes a built-in CLI for quick testing and common tasks without writing custom Python code.