-
Hi there,
First off, amazing job on your paper/the model! It looks super promising.
I'm working on a project where I'm attempting to do live streaming with Whisper. One of the challenges there i…
-
I noticed you're transcribing the video by sending a request to your server:
https://github.com/paprikka/butter/blob/15456f8ca583c3b537e84ed770950698594fd0a6/src/watcher/get-transcript.ts#L4-L19
…
-
### Brief Description
In the terminate() function within vocode/streaming/streaming_conversation.py, after the log entry `Terminating events Task`, the rest of the function does not execute. The lo…
-
## Summary
The task is to create an article presenting a transcription and analysis of an episode of the podcast 'Episode 04 — Tokens', with participation from Kirill Emelyanenko from TON Core.
…
-
File "F:\my_AI\transcribe-main\app\transcribe\db\app_db.py", line 10, in
from tsutils import Singleton # noqa: E402 pylint: disable=C0413
ImportError: cannot import name 'Singleton' from 'ts…
-
- Telegram allows group voice/video calls to be recorded.
- When the call is finished, Telegram saves the recording in "saved messages"
- It is VERY fast to forward that content to the bot.
- The …
-
Custom language model is available for use on AWS API but is not available in Jambonz. We are currently using Custom Vocabulary but we are finding it difficult with its limitaions as max size of vocab…
-
The language "zh" is not supported by the model "long" in the location named "global".
-
### System Info
xenova/transformers.js#v3
### Environment/Platform
- [ ] Website/web-app
- [ ] Browser extension
- [ ] Server-side (e.g., Node.js, Deno, Bun)
- [ ] Desktop app (e.g., Elect…
-
Currently batch processing attempts to process all files in the given directory, even hidden ones (macOS's `.DS_store` is currently deny listed). Attempting to process a non-audio file causes the tran…