Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

stenographer interface #175

Open
wants to merge 23 commits into
base: master
Choose a base branch
from
Open

stenographer interface #175

wants to merge 23 commits into from

Conversation

royshil
Copy link
Collaborator

@royshil royshil commented Oct 9, 2024

No description provided.

This commit adds the following changes:
- Added a new section for stenographer options in the filter properties
- Implemented resample utilities for handling audio data

These changes enable the use of stenographer functionality and provide support for resampling audio data.
Added a new option for stenographer delay in the stenographer group properties. This allows users to specify the amount of delay for partial transcription. The default delay is set to 10,000 milliseconds.

Fixes #<issue_number>
This commit adds the file resample-utils.cpp to the whisper-utils directory. It also updates the stenographer delay functionality in stenographer.cpp and transcription-filter.cpp. The delay is now set to 1000 milliseconds. These changes improve the audio processing and transcription capabilities of the application.
Update the variable name from "stenographer_delay" to "stenographer_delay_ms" in the transcription filter code. This change reflects the unit of the delay value in milliseconds. The code has been modified in the "transcription-filter-data.h" and "transcription-filter.cpp" files.
This commit modifies the `reset_caption_state` function in `transcription-filter-callbacks.cpp` to clear the `stenographer_delay_buffers` when resetting the caption state. This ensures that the buffers are empty and ready for new data.

Additionally, the `channels` variable in `transcription-filter-data.h` is updated to represent the number of channels in the input.

These changes improve the functionality and maintainability of the code.
This commit refactors the CMakeLists.txt file by adding a check for the buildspec variable and reading the buildspec.json file if it is not set. It also sets the arch variable based on the platform. In stenographer.cpp, the code now includes websocketpp/config/asio_no_tls.hpp only if the platform is not Linux. Additionally, the unused parameter hdl is now marked as unused in the message handler.
…ranslation context

- Remove commented code and improve detokenization logic in translation.cpp
- Add support for translation context by adding last input tokens and last translation tokens
- Add option to remove punctuation from the start of translations
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant