stream-translator-gpt 2026.5.20

Command line tool to transcribe & translate audio from livestreams in real time

stream-translator-gpt

stream-translator-gpt is a command-line tool for real-time transcription and translation of live streams. We have now added an easier-to-use WebUI entry point.

Try it on Colab:

WebUI	Command Line

(Due to frequent scraping and theft of API keys, we are unable to provide a trial API key. You need to fill in your own API key.)

Prerequisites

1. Python >= 3.8 (Recommend >= 3.10)
2. FFmpeg (skip if already installed):
   • Windows: winget install ffmpeg
   • Linux (Debian/Ubuntu): sudo apt install ffmpeg
3. Install CUDA on your system.
4. Install cuDNN to your CUDA dir if you want to use Faster-Whisper.
5. Install PyTorch (with CUDA) to your Python.
6. Create a Google API key if you want to use Gemini API for translation.
7. Create a OpenAI API key if you want to use OpenAI Transcription API for transcription or GPT API for translation.

WebUI

pip install stream-translator-gpt[webui] -U
stream-translator-gpt-webui

Command Line

Install release version from PyPI:

pip install stream-translator-gpt -U
stream-translator-gpt

or

Clone master version code from Github:

git clone https://github.com/ionic-bond/stream-translator-gpt.git
pip install -r ./stream-translator-gpt/requirements.txt -U
python3 ./stream-translator-gpt/stream_translator_gpt/main.py

Usage

The commands on Colab are the recommended usage, below are some other commonly used options.

• Transcribe live streaming (default use Whisper):

  stream-translator-gpt {URL} --model large --language {input_language}

• Transcribe by Faster-Whisper:

  stream-translator-gpt {URL} --model large --language {input_language} --use_faster_whisper

• Transcribe by SimulStreaming:

  stream-translator-gpt {URL} --model large --language {input_language} --use_simul_streaming

• Transcribe by SimulStreaming with Faster-Whisper as the encoder:

  stream-translator-gpt {URL} --model large --language {input_language} --use_simul_streaming --use_faster_whisper

• Transcribe by OpenAI Transcription API:

  stream-translator-gpt {URL} --language {input_language} --use_openai_transcription_api --openai_api_key {your_openai_key}

• Transcribe by a HuggingFace ASR model (requires pip install stream-translator-gpt[hf_asr]):

  stream-translator-gpt {URL} --model {hf_model_name} --use_hf_asr

  Only models with pipeline_tag: automatic-speech-recognition on Hugging Face Hub are supported.

• Translate to other language by Gemini:

  stream-translator-gpt {URL} --model large --language ja --translation_prompt "Translate from Japanese to Chinese" --google_api_key {your_google_key}

• Translate to other language by GPT:

  stream-translator-gpt {URL} --model large --language ja --translation_prompt "Translate from Japanese to Chinese" --openai_api_key {your_openai_key}

• Using OpenAI Transcription API and Gemini at the same time:

  stream-translator-gpt {URL} --language ja --use_openai_transcription_api --openai_api_key {your_openai_key} --translation_prompt "Translate from Japanese to Chinese" --google_api_key {your_google_key}

• Local video/audio file as input:

  stream-translator-gpt /path/to/file --model large --language {input_language}

• Record system audio as input:

  stream-translator-gpt device --model large --language {input_language}

• Record microphone as input:

  stream-translator-gpt device --model large --language {input_language} --mic

• Sending result to Discord:

  stream-translator-gpt {URL} --model large --language {input_language} --discord_webhook_url {your_discord_webhook_url}

• Sending result to Telegram:

  stream-translator-gpt {URL} --model large --language {input_language} --telegram_token {your_telegram_token} --telegram_chat_id {your_telegram_chat_id}

• Sending result to Cqhttp:

  stream-translator-gpt {URL} --model large --language {input_language} --cqhttp_url {your_cqhttp_url} --cqhttp_token {your_cqhttp_token}

• Saving result to a .srt subtitle file:

  stream-translator-gpt {URL} --model large --language ja --translation_prompt "Translate from Japanese to Chinese" --google_api_key {your_google_key} --hide_transcribe_result --retry_if_translation_fails --output_timestamps --output_file_path ./result.srt