From a6d0a062beab5191ce15e7da242d5a283f64743e Mon Sep 17 00:00:00 2001 From: pabloFuente Date: Wed, 2 Jul 2025 20:15:00 +0200 Subject: [PATCH] Updated agent-speech-processing.yaml with Spitch provider --- community/agent-speech-processing.yaml | 14 ++++++++++---- pro/agent-speech-processing.yaml | 14 ++++++++++---- 2 files changed, 20 insertions(+), 8 deletions(-) diff --git a/community/agent-speech-processing.yaml b/community/agent-speech-processing.yaml index 756390f..3375c79 100644 --- a/community/agent-speech-processing.yaml +++ b/community/agent-speech-processing.yaml @@ -10,7 +10,7 @@ live_captions: # - manual: the agent will connect to new Rooms only when your application dictates it by using the Agent Dispatch API. processing: automatic - # Which speech-to-text AI provider to use [aws, azure, google, opeanai, groq, deepgram, assemblyai, fal, clova, speechmatics, gladia, sarvam] + # Which speech-to-text AI provider to use [aws, azure, google, openai, groq, deepgram, assemblyai, fal, clova, speechmatics, gladia, sarvam, spitch] # The custom configuration for the selected provider must be set below provider: @@ -154,8 +154,8 @@ live_captions: detect_language: false # Whether to return interim (non-final) transcription results. Defaults to true. See https://developers.deepgram.com/docs/interim-results interim_results: true - # Whether to apply smart formatting to numbers, dates, etc. Defaults to true. See https://developers.deepgram.com/docs/smart-format - smart_format: true + # Whether to apply smart formatting to numbers, dates, etc. Defaults to false. See https://developers.deepgram.com/docs/smart-format + smart_format: false # When smart_format is used, ensures it does not wait for sequence to be complete before returning results. Defaults to true. See https://developers.deepgram.com/docs/smart-format#using-no-delay no_delay: true # Whether to add punctuations to the transcription. Defaults to true. Turn detector will work better with punctuations. See https://developers.deepgram.com/docs/punctuation @@ -222,7 +222,7 @@ live_captions: speaker_sensitivity: # See https://docs.speechmatics.com/features/diarization#prefer-current-speaker prefer_current_speaker: - # Permitted puctuation marks for advanced punctuation. See https://docs.speechmatics.com/features/punctuation-settings + # Permitted punctuation marks for advanced punctuation. See https://docs.speechmatics.com/features/punctuation-settings # Commented is an example of punctuation settings punctuation_overrides: # permitted_marks: [ ".", "," ] @@ -257,3 +257,9 @@ live_captions: language: # The Sarvam STT model to use. See https://docs.sarvam.ai/api-reference-docs/speech-to-text/transcribe#request.body.model.model model: + + spitch: + # API key for Spitch. See https://www.spitch.studio/api-keys + api_key: + # See https://docs.spitch.app/features/transcription#language-support + language: diff --git a/pro/agent-speech-processing.yaml b/pro/agent-speech-processing.yaml index 756390f..c68d354 100644 --- a/pro/agent-speech-processing.yaml +++ b/pro/agent-speech-processing.yaml @@ -10,7 +10,7 @@ live_captions: # - manual: the agent will connect to new Rooms only when your application dictates it by using the Agent Dispatch API. processing: automatic - # Which speech-to-text AI provider to use [aws, azure, google, opeanai, groq, deepgram, assemblyai, fal, clova, speechmatics, gladia, sarvam] + # Which speech-to-text AI provider to use [aws, azure, google, openai, groq, deepgram, assemblyai, fal, clova, speechmatics, gladia, sarvam, spitch] # The custom configuration for the selected provider must be set below provider: @@ -154,8 +154,8 @@ live_captions: detect_language: false # Whether to return interim (non-final) transcription results. Defaults to true. See https://developers.deepgram.com/docs/interim-results interim_results: true - # Whether to apply smart formatting to numbers, dates, etc. Defaults to true. See https://developers.deepgram.com/docs/smart-format - smart_format: true + # Whether to apply smart formatting to numbers, dates, etc. Defaults to false. See https://developers.deepgram.com/docs/smart-format + smart_format: false # When smart_format is used, ensures it does not wait for sequence to be complete before returning results. Defaults to true. See https://developers.deepgram.com/docs/smart-format#using-no-delay no_delay: true # Whether to add punctuations to the transcription. Defaults to true. Turn detector will work better with punctuations. See https://developers.deepgram.com/docs/punctuation @@ -222,7 +222,7 @@ live_captions: speaker_sensitivity: # See https://docs.speechmatics.com/features/diarization#prefer-current-speaker prefer_current_speaker: - # Permitted puctuation marks for advanced punctuation. See https://docs.speechmatics.com/features/punctuation-settings + # Permitted punctuation marks for advanced punctuation. See https://docs.speechmatics.com/features/punctuation-settings # Commented is an example of punctuation settings punctuation_overrides: # permitted_marks: [ ".", "," ] @@ -257,3 +257,9 @@ live_captions: language: # The Sarvam STT model to use. See https://docs.sarvam.ai/api-reference-docs/speech-to-text/transcribe#request.body.model.model model: + + spitch: + # API key for Spitch. See https://www.spitch.studio/api-keys + api_key: + # See https://docs.spitch.app/features/transcription#language-support + language: \ No newline at end of file