Skip to content

chore: Adds TranscriptFilteringConfig and ConversationConfig to batch and realtime models#21

Open
shubham-chawla wants to merge 5 commits into
speechmatics:mainfrom
shubham-chawla:main
Open

chore: Adds TranscriptFilteringConfig and ConversationConfig to batch and realtime models#21
shubham-chawla wants to merge 5 commits into
speechmatics:mainfrom
shubham-chawla:main

Conversation

@shubham-chawla

@shubham-chawla shubham-chawla commented Dec 23, 2025

Copy link
Copy Markdown

Adds TranscriptFilteringConfig
Usage:

let mut config: SessionConfig = Default::default();

config.transcription_config.transcript_filtering_config = Some(Box::new(models::TranscriptFilteringConfig {
  remove_disfluencies: Some(true),
}));

Adds ConversationConfig
Usage:

let mut config: SessionConfig = Default::default();

config.transcription_config.conversation_config = Some(Box::new(models::ConversationConfig {
  end_of_utterance_silence_trigger: Some(1.2_f32),
}));

@shubham-chawla shubham-chawla changed the title Adds TranscriptFilteringConfig to batch and real Adds TranscriptFilteringConfig to batch and real models Dec 23, 2025
@shubham-chawla shubham-chawla changed the title Adds TranscriptFilteringConfig to batch and real models chore: Adds TranscriptFilteringConfig to batch and real models Dec 23, 2025
@shubham-chawla

Copy link
Copy Markdown
Author

@TudorCRL Could you please take a look at this PR?

@shubham-chawla shubham-chawla changed the title chore: Adds TranscriptFilteringConfig to batch and real models chore: Adds TranscriptFilteringConfig to batch and realtime models Dec 25, 2025
@TudorCRL

Copy link
Copy Markdown
Collaborator

@shubham-chawla apologies for the delayed responsed, I missed during the holidays and I've been ill this week! Your PR looks great. Thanks for raising it! I've approved it. Also, more generally, let me know if you have any feedback on the SDK and things we could do to improve it.

@shubham-chawla

shubham-chawla commented Jan 16, 2026

Copy link
Copy Markdown
Author

@shubham-chawla apologies for the delayed responsed, I missed during the holidays and I've been ill this week! Your PR looks great. Thanks for raising it! I've approved it. Also, more generally, let me know if you have any feedback on the SDK and things we could do to improve it.

Thanks @TudorCRL! I hope you are feeling better.
I think the SDK would benefit from additional examples. At the moment, it only includes a single example using a sample file. It would be helpful to add more examples, especially around using the microphone stream.

@shubham-chawla shubham-chawla changed the title chore: Adds TranscriptFilteringConfig to batch and realtime models chore: Adds TranscriptFilteringConfig and ConversationConfig to batch and realtime models Feb 18, 2026
@TudorCRL

Copy link
Copy Markdown
Collaborator

@shubham-chawla apologies for the delayed responsed, I missed during the holidays and I've been ill this week! Your PR looks great. Thanks for raising it! I've approved it. Also, more generally, let me know if you have any feedback on the SDK and things we could do to improve it.

Thanks @TudorCRL! I hope you are feeling better. I think the SDK would benefit from additional examples. At the moment, it only includes a single example using a sample file. It would be helpful to add more examples, especially around using the microphone stream.

Hey @shubham-chawla, thanks, I'm better now! This is good feedback re having more examples. The challenge we have is that our rust SDK isn't super widely used so isn't our primary priority, and I'm not sure this will change in the near term. Also, it might be of interest to you that we've launched the Speechmatics Academy, which has a wider set of use case-focussed code examples (currently all in python but soon to be expanded to other languages). Idk if they'd be super helpful for you in particular as they don't have rust, but might be worth checking out.

Btw for this PR, I've approved the pipeline but it looks like there's a test failure somewhere.

@shubham-chawla

Copy link
Copy Markdown
Author

@shubham-chawla apologies for the delayed responsed, I missed during the holidays and I've been ill this week! Your PR looks great. Thanks for raising it! I've approved it. Also, more generally, let me know if you have any feedback on the SDK and things we could do to improve it.

Thanks @TudorCRL! I hope you are feeling better. I think the SDK would benefit from additional examples. At the moment, it only includes a single example using a sample file. It would be helpful to add more examples, especially around using the microphone stream.

Hey @shubham-chawla, thanks, I'm better now! This is good feedback re having more examples. The challenge we have is that our rust SDK isn't super widely used so isn't our primary priority, and I'm not sure this will change in the near term. Also, it might be of interest to you that we've launched the Speechmatics Academy, which has a wider set of use case-focussed code examples (currently all in python but soon to be expanded to other languages). Idk if they'd be super helpful for you in particular as they don't have rust, but might be worth checking out.

Btw for this PR, I've approved the pipeline but it looks like there's a test failure somewhere.

Thanks @TudorCRL ! I checked the pipline and it says

Text("{\"message\":\"Error\",\"type\":\"not_authorised\",\"reason\":\"Not Authorized\"}")

I think some env var is missing from the workflow.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Development

Successfully merging this pull request may close these issues.

2 participants