(POC, School project, LIA)
The program can create a transcription from an audio (.mp4) or video file (.wav). It also performs speaker diarization of the conversation. This is particularly useful for recording meetings in Teams. Additionally, it can provide a summary or analysis of the conversation. The conversation, along with the summary and analysis, can be stored in the program, and old stored conversations can also be deleted. It is a Streamlit application.
-
Transcribe audio from files (wav, mp4) or YouTube URLs.
-
Perform speaker diarization to identify and separate different speakers in the audio.
-
Display full conversation transcripts.
-
Summarize and analyze transcriptions using OpenAI GPT.
-
Save, load, and delete transcriptions.
-
Send transcriptions and summaries via email.
Python version required: 3.10
Libraries in requirements.txt: ffmpeg whisper pyannote.audio streamlit openai==0.28.0 yt_dlp
You can install these packages by running the following command in your terminal:
pip install -r requirements.txt
-
The OpenAI API key (OPENAI_API_KEY) needs to be bought and it is recommended to set this key in the environmental variables.
-
The diarization key (AUTH_TOKEN_VOCATIO) from Hugging Face needs to be set and it is recommended to set this key in the environmental variables.
-
The email address (SMTP_EMAIL_USER) needs to be set and it is recommended to set this key in the environmental variables.
-
The email password (SMTP_EMAIL_PASSWORD) needs to be set and it is recommended to set this key in the environmental variables.
-
You can set a environmental key like this in windows in your project folder: setx OPENAI_API_KEY ".....key_value......"
streamlit run app.py