WhisperLive
WhisperLive copied to clipboard
collabora
A nearly-live implementation of OpenAI's Whisper.
When the speech time is long and words are continuous, delay goes little bit higher. The model runs on Nvidia T4 GPU. Should I consider upgrading my GPU model or...
Hello, First of all, thank you a lot for this library. I wonder; is it possible to obtain word-level timestamps from this model like it is from WhisperX, whisper-timestamped, faster-whisper...
I have attached my client and server code. I have saved the client details in a py file and am calling it. I am using faster_whisper backend without tensorrt. Can...
I'm running the CPU version of Whisper Live, and here's the client code I'm using: from whisper_live.client import TranscriptionClient client = TranscriptionClient( "localhost", 9090, lang="en", translate=False # Only used for...
 Something is wrong with the server.
Hi, I have built a TensoRT engine and tried running the command: ``` python3 run_server.py -p 9090 -b tensorrt -trt {path_to_engine} ``` but the only output that I have received...
No need to bother with run_server script. After package installation the server can be run using `python -m whisper_live.server` (similar to `python -m http.server`)
I kept encountering hallucinations when I tried transcribe from microphone, if "lang=en" it will be "Thank you" , if "lang=zh" it will be something like"字幕by索兰娅" "字幕由 Amara.org 社群提供", I think...
` from whisper_live.client import TranscriptionClient client = TranscriptionClient( "localhost", 9090, lang="zh", translate=False, model="large-v3", use_vad=False, save_output_recording=False, # Only used for microphone input, False by Default output_recording_filename="./output_recording.wav", # Only used for microphone...
