assemblyai-transcribe

Small CLI for uploading local audio file(s) to AssemblyAI and saving the transcript(s).

The code is compatible with Python 2.7 and Python 3.x.

Installation

Install from PyPI:

pip install assemblyai-transcribe

After installation, run it as assemblyai_transcribe.

Usage

Transcribe a single file:

assemblyai_transcribe ./example.mp3 --api-key YOUR_API_KEY -o transcript.txt

Transcribe multiple files:

assemblyai_transcribe ./ep1.mp3 ./ep2.mp3 ./ep3.mp3 --api-key YOUR_API_KEY
# Saves: ep1.mp3.txt, ep2.mp3.txt, ep3.mp3.txt

Transcribe multiple files into a directory (auto-created if it doesn't exist):

assemblyai_transcribe ./ep1.mp3 ./ep2.mp3 --api-key YOUR_API_KEY -o transcripts/
# Saves: transcripts/ep1.txt, transcripts/ep2.txt

Examples:

assemblyai_transcribe ./example.mp3 --api-key YOUR_API_KEY -o transcript.txt
assemblyai_transcribe ./example.mp3 --api-key YOUR_API_KEY --eu -o transcript.txt
assemblyai_transcribe ./example.mp3 --api-key YOUR_API_KEY --model universal-3-pro -o transcript.txt
assemblyai_transcribe ./example.mp3 --api-key YOUR_API_KEY --model universal-3-pro --model universal-2 -o transcript.txt
assemblyai_transcribe ./example.mp3 --api-key YOUR_API_KEY --no-language-detection -o transcript.txt
assemblyai_transcribe ./example.mp3 --api-key YOUR_API_KEY --no-speaker-labels -o transcript.txt

Arguments:

• audio_files: one or more paths to local audio files.
• --api-key API_KEY: AssemblyAI API key.
• --base-url BASE_URL: override the API base URL. Default: https://api.assemblyai.com
• --eu: use the EU AssemblyAI endpoint.
• --model MODEL: repeat to provide multiple speech models. Default models: universal-3-pro, universal-2
• --no-speaker-labels: disable speaker diarization (enabled by default).
• --no-language-detection: disable automatic language detection (enabled by default).
• --poll-interval SECONDS: seconds between polling attempts. Default: 3.0
• -o OUTPUT, --output OUTPUT: write transcript(s). With one input file, this is the output file path (default: AUDIO_FILE.txt). With multiple input files, this must be a directory; each transcript is saved as BASENAME.txt inside it.

Help:

assemblyai_transcribe --help

Contributing

Contributions are welcome! Please submit pull requests or open issues on the GitHub repository.

License

This project is licensed under the MIT License.