Carlos Ruiz Ballesteros
GitHub
commit
2ae40b3a40
@ -1,5 +1,5 @@
|
||||
# Docker image of the agent.
|
||||
docker_image: docker.io/openvidu/agent-speech-processing:3.4.1
|
||||
docker_image: docker.io/openvidu/agent-speech-processing:3.5.0
|
||||
|
||||
# Whether to run the agent or not.
|
||||
enabled: false
|
||||
@ -16,7 +16,7 @@ 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, openai, groq, deepgram, assemblyai, fal, clova, speechmatics, gladia, sarvam, spitch]
|
||||
# Which speech-to-text AI provider to use [aws, azure, google, openai, azure_openai, groq, deepgram, assemblyai, fal, clova, speechmatics, gladia, sarvam, mistralai, cartesia, soniox]
|
||||
# The custom configuration for the selected provider must be set below
|
||||
provider:
|
||||
|
||||
@ -63,6 +63,10 @@ live_captions:
|
||||
# Removes profanity (swearing), or replaces letters of profane words with stars. Valid values: Masked | Removed | Raw
|
||||
# See https://learn.microsoft.com/en-us/azure/ai-services/translator/profanity-filtering
|
||||
profanity:
|
||||
# List of words or phrases to boost recognition accuracy. Azure will give higher priority to these phrases during recognition.
|
||||
phrase_list:
|
||||
# Controls punctuation behavior. If True, enables explicit punctuation mode where punctuation marks are added explicitly. If False (default), uses Azure's default punctuation behavior.
|
||||
explicit_punctuation:
|
||||
|
||||
azure_openai:
|
||||
# Credentials for Azure OpenAI APIs. See https://learn.microsoft.com/en-us/azure/api-management/api-management-authenticate-authorize-azure-openai
|
||||
@ -82,6 +86,8 @@ live_captions:
|
||||
project:
|
||||
# The language code to use for transcription (e.g., "en" for English).
|
||||
language:
|
||||
# Whether to automatically detect the language.
|
||||
detect_language:
|
||||
# ID of the model to use for speech-to-text.
|
||||
model:
|
||||
# Initial prompt to guide the transcription.
|
||||
@ -135,6 +141,8 @@ live_captions:
|
||||
# The language of the input audio. Supplying the input language in ISO-639-1 format
|
||||
# (https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) will improve accuracy and latency.
|
||||
language:
|
||||
# Whether to automatically detect the language.
|
||||
detect_language:
|
||||
# Optional text prompt to guide the transcription. Only supported for whisper-1.
|
||||
prompt:
|
||||
|
||||
@ -146,8 +154,12 @@ live_captions:
|
||||
# The language of the input audio. Supplying the input language in ISO-639-1 format
|
||||
# (https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) will improve accuracy and latency.
|
||||
language:
|
||||
# Whether to automatically detect the language.
|
||||
detect_language:
|
||||
# Prompt to guide the model's style or specify how to spell unfamiliar words. 224 tokens max.
|
||||
prompt:
|
||||
# Base URL for the Groq API. By default "https://api.groq.com/openai/v1"
|
||||
base_url:
|
||||
|
||||
deepgram:
|
||||
# See https://console.deepgram.com/
|
||||
@ -156,25 +168,27 @@ live_captions:
|
||||
model:
|
||||
# See https://developers.deepgram.com/reference/speech-to-text-api/listen-streaming#request.query.language
|
||||
language:
|
||||
# Whether to enable automatic language detection. Defaults to false. See https://developers.deepgram.com/docs/language-detection
|
||||
# Whether to enable automatic language detection. See https://developers.deepgram.com/docs/language-detection
|
||||
detect_language: false
|
||||
# Whether to return interim (non-final) transcription results. Defaults to true. See https://developers.deepgram.com/docs/interim-results
|
||||
# Whether to return interim (non-final) transcription results. See https://developers.deepgram.com/docs/interim-results
|
||||
interim_results: true
|
||||
# Whether to apply smart formatting to numbers, dates, etc. Defaults to false. See https://developers.deepgram.com/docs/smart-format
|
||||
# Whether to apply smart formatting to numbers, dates, etc. 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
|
||||
# When smart_format is used, ensures it does not wait for sequence to be complete before returning results. 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
|
||||
# Whether to add punctuations to the transcription. Turn detector will work better with punctuations. See https://developers.deepgram.com/docs/punctuation
|
||||
punctuate: true
|
||||
# Whether to include filler words (um, uh, etc.) in transcription. Defaults to true. See https://developers.deepgram.com/docs/filler-words
|
||||
# Whether to include filler words (um, uh, etc.) in transcription. See https://developers.deepgram.com/docs/filler-words
|
||||
filler_words: true
|
||||
# Whether to filter profanity from the transcription. Defaults to false. See https://developers.deepgram.com/docs/profanity-filter
|
||||
# Whether to filter profanity from the transcription. See https://developers.deepgram.com/docs/profanity-filter
|
||||
profanity_filter: false
|
||||
# List of tuples containing keywords and their boost values for improved recognition. Each tuple should be (keyword: str, boost: float). Defaults to None. keywords does not work with Nova-3 models. Use keyterms instead.
|
||||
# Whether to transcribe numbers as numerals. See https://developers.deepgram.com/docs/numerals
|
||||
numerals: false
|
||||
# List of tuples containing keywords and their boost values for improved recognition. Each tuple should be (keyword: str, boost: float). keywords does not work with Nova-3 models. Use keyterms instead.
|
||||
# keywords:
|
||||
# - [OpenVidu, 1.5]
|
||||
# - [WebRTC, 1]
|
||||
# List of key terms to improve recognition accuracy. Defaults to None. keyterms is supported by Nova-3 models.
|
||||
# List of key terms to improve recognition accuracy. keyterms is supported by Nova-3 models.
|
||||
# Commented below is an example
|
||||
keyterms:
|
||||
# - "OpenVidu"
|
||||
@ -183,8 +197,18 @@ live_captions:
|
||||
assemblyai:
|
||||
# API key for AssemblyAI. See https://www.assemblyai.com/dashboard/api-keys
|
||||
api_key:
|
||||
# The confidence threshold (0.0 to 1.0) to use when determining if the end of a turn has been reached.
|
||||
end_of_turn_confidence_threshold:
|
||||
# The minimum amount of silence in milliseconds required to detect end of turn when confident.
|
||||
min_end_of_turn_silence_when_confident:
|
||||
# The maximum amount of silence in milliseconds allowed in a turn before end of turn is triggered.
|
||||
max_turn_silence:
|
||||
# Whether to return formatted final transcripts (proper punctuation, letter casing...). If enabled, formatted final transcripts are emitted shortly following an end-of-turn detection.
|
||||
format_turns: true
|
||||
# List of keyterms to improve recognition accuracy for specific words and phrases.
|
||||
keyterms_prompt:
|
||||
# - "OpenVidu"
|
||||
# - "WebRTC"
|
||||
|
||||
fal:
|
||||
# API key for fal. See https://fal.ai/dashboard/keys
|
||||
@ -208,12 +232,14 @@ live_captions:
|
||||
speechmatics:
|
||||
# API key for Speechmatics. See https://portal.speechmatics.com/manage-access/
|
||||
api_key:
|
||||
# ISO 639-1 language code. All languages are global and can understand different dialects/accents. To see the list of all supported languages, see https://docs.speechmatics.com/introduction/supported-languages
|
||||
# ISO 639-1 language code. All languages are global and can understand different dialects/accents. To see the list of all supported languages, see https://docs.speechmatics.com/speech-to-text/languages#transcription-languages
|
||||
language:
|
||||
# Operating point to use for the transcription per required accuracy & complexity. To learn more, see https://docs.speechmatics.com/features/accuracy-language-packs#accuracy
|
||||
# Operating point to use for the transcription per required accuracy & complexity. To learn more, see https://docs.speechmatics.com/speech-to-text/languages#operating-points
|
||||
operating_point:
|
||||
# Partial transcripts allow you to receive preliminary transcriptions and update as more context is available until the higher-accuracy final transcript is returned. Partials are returned faster but without any post-processing such as formatting. See https://docs.speechmatics.com/features/realtime-latency#partial-transcripts
|
||||
# Partial transcripts allow you to receive preliminary transcriptions and update as more context is available until the higher-accuracy final transcript is returned. Partials are returned faster but without any post-processing such as formatting. See https://docs.speechmatics.com/speech-to-text/realtime/output#partial-transcripts
|
||||
enable_partials:
|
||||
# Enable speaker diarization. When enabled, the STT engine will determine and attribute words to unique speakers. The speaker_sensitivity parameter can be used to adjust the sensitivity of diarization
|
||||
enable_diarization:
|
||||
# RFC-5646 language code to make spelling rules more consistent in the transcription output. See https://docs.speechmatics.com/features/word-tagging#output-locale
|
||||
output_locale:
|
||||
# The delay in seconds between the end of a spoken word and returning the final transcript results. See https://docs.speechmatics.com/features/realtime-latency#configuration-example
|
||||
@ -255,6 +281,10 @@ live_captions:
|
||||
languages:
|
||||
# Whether to allow switching between languages during recognition. Defaults to True
|
||||
code_switching:
|
||||
# https://docs.gladia.io/api-reference/v2/live/init#body-pre-processing-audio-enhancer
|
||||
pre_processing_audio_enhancer:
|
||||
# https://docs.gladia.io/api-reference/v2/live/init#body-pre-processing-speech-threshold
|
||||
pre_processing_speech_threshold:
|
||||
|
||||
sarvam:
|
||||
# API key for Sarvam. See https://dashboard.sarvam.ai/key-management
|
||||
@ -263,3 +293,29 @@ 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:
|
||||
|
||||
mistralai:
|
||||
# API key for Mistral AI. See https://console.mistral.ai/api-keys
|
||||
api_key:
|
||||
# Name of the Voxtral STT model to use. Default to voxtral-mini-latest. See https://docs.mistral.ai/capabilities/audio/
|
||||
model:
|
||||
# The language code to use for transcription (e.g., "en" for English)
|
||||
language:
|
||||
|
||||
cartesia:
|
||||
# API key for Cartesia. See https://play.cartesia.ai/keys
|
||||
api_key:
|
||||
# The Cartesia STT model to use
|
||||
model:
|
||||
# The language code to use for transcription (e.g., "en" for English)
|
||||
language:
|
||||
|
||||
soniox:
|
||||
# API key for Soniox. See https://console.soniox.com/
|
||||
api_key:
|
||||
# Set language hints when possible to significantly improve accuracy. See: https://soniox.com/docs/stt/concepts/language-hints
|
||||
language_hints:
|
||||
# - "en"
|
||||
# - "es"
|
||||
# Set context to improve recognition of difficult and rare words. Context is a string and can include words, phrases, sentences, or summaries (limit: 10K chars). See https://soniox.com/docs/stt/concepts/context
|
||||
context:
|
||||
@ -14,14 +14,4 @@ if [ -z "$LAN_PRIVATE_IP" ]; then
|
||||
fi
|
||||
|
||||
# Replace the LAN_PRIVATE_IP in the .env file
|
||||
sed -i'' -e "s/LAN_PRIVATE_IP=.*/LAN_PRIVATE_IP=$LAN_PRIVATE_IP/g" .env
|
||||
|
||||
# If sillicon mac, enable EXPERIMENTAL_DOCKER_DESKTOP_FORCE_QEMU flag
|
||||
if [ "$(uname -m)" = "arm64" ]; then
|
||||
if ! grep -q "EXPERIMENTAL_DOCKER_DESKTOP_FORCE_QEMU" .env; then
|
||||
echo "# Enable this flag to run Docker Desktop on Apple Silicon Macs" >> .env
|
||||
echo "EXPERIMENTAL_DOCKER_DESKTOP_FORCE_QEMU=1" >> .env
|
||||
else
|
||||
sed -i'' -e "s/EXPERIMENTAL_DOCKER_DESKTOP_FORCE_QEMU=.*/EXPERIMENTAL_DOCKER_DESKTOP_FORCE_QEMU=1/g" .env
|
||||
fi
|
||||
fi
|
||||
sed -i'' -e "s/LAN_PRIVATE_IP=.*/LAN_PRIVATE_IP=$LAN_PRIVATE_IP/g" .env
|
||||
@ -1,7 +1,6 @@
|
||||
services:
|
||||
caddy-proxy:
|
||||
image: docker.io/openvidu/openvidu-caddy-local:3.4.1
|
||||
platform: linux/amd64
|
||||
image: docker.io/openvidu/openvidu-caddy-local:3.5.0
|
||||
container_name: caddy-proxy
|
||||
restart: unless-stopped
|
||||
extra_hosts:
|
||||
@ -22,6 +21,7 @@ services:
|
||||
- MEET_INITIAL_API_KEY=${MEET_INITIAL_API_KEY:-}
|
||||
volumes:
|
||||
- ./custom-layout:/var/www/custom-layout
|
||||
- /etc/localtime:/etc/localtime:ro
|
||||
ports:
|
||||
- 5443:5443
|
||||
- 6443:6443
|
||||
@ -33,14 +33,14 @@ services:
|
||||
condition: service_completed_successfully
|
||||
|
||||
redis:
|
||||
image: docker.io/redis:7.4.4-alpine
|
||||
platform: linux/amd64
|
||||
image: docker.io/redis:8.2.2-alpine
|
||||
container_name: redis
|
||||
restart: unless-stopped
|
||||
ports:
|
||||
- 6379:6379
|
||||
volumes:
|
||||
- redis:/data
|
||||
- /etc/localtime:/etc/localtime:ro
|
||||
command: >
|
||||
redis-server
|
||||
--bind 0.0.0.0
|
||||
@ -50,8 +50,7 @@ services:
|
||||
condition: service_completed_successfully
|
||||
|
||||
minio:
|
||||
image: docker.io/openvidu/minio:2025.5.24-debian-12-r1
|
||||
platform: linux/amd64
|
||||
image: docker.io/openvidu/minio:2025.9.7-debian-12-r3
|
||||
container_name: minio
|
||||
restart: unless-stopped
|
||||
ports:
|
||||
@ -61,23 +60,25 @@ services:
|
||||
- MINIO_ROOT_PASSWORD=${MINIO_SECRET_KEY:-}
|
||||
- MINIO_DEFAULT_BUCKETS=openvidu-appdata
|
||||
- MINIO_CONSOLE_SUBPATH=/minio-console
|
||||
- MINIO_BROWSER=on
|
||||
- MINIO_BROWSER_REDIRECT_URL=http://localhost:7880/minio-console
|
||||
volumes:
|
||||
- minio-data:/bitnami/minio/data
|
||||
- minio-certs:/certs
|
||||
- /etc/localtime:/etc/localtime:ro
|
||||
depends_on:
|
||||
setup:
|
||||
condition: service_completed_successfully
|
||||
|
||||
mongo:
|
||||
image: docker.io/openvidu/mongodb:8.0.12-r0
|
||||
platform: linux/amd64
|
||||
image: docker.io/openvidu/mongodb:8.0.15-r0
|
||||
container_name: mongo
|
||||
restart: unless-stopped
|
||||
ports:
|
||||
- 27017:27017
|
||||
volumes:
|
||||
- mongo-data:/bitnami/mongodb
|
||||
- /etc/localtime:/etc/localtime:ro
|
||||
environment:
|
||||
- MONGODB_ROOT_USER=${MONGO_ADMIN_USERNAME:-}
|
||||
- MONGODB_ROOT_PASSWORD=${MONGO_ADMIN_PASSWORD:-}
|
||||
@ -85,14 +86,12 @@ services:
|
||||
- MONGODB_REPLICA_SET_MODE=primary
|
||||
- MONGODB_REPLICA_SET_NAME=rs0
|
||||
- MONGODB_REPLICA_SET_KEY=devreplicasetkey
|
||||
- EXPERIMENTAL_DOCKER_DESKTOP_FORCE_QEMU=${EXPERIMENTAL_DOCKER_DESKTOP_FORCE_QEMU:-0}
|
||||
depends_on:
|
||||
setup:
|
||||
condition: service_completed_successfully
|
||||
|
||||
dashboard:
|
||||
image: docker.io/openvidu/openvidu-dashboard:3.4.1
|
||||
platform: linux/amd64
|
||||
image: docker.io/openvidu/openvidu-dashboard:3.5.0
|
||||
container_name: dashboard
|
||||
restart: unless-stopped
|
||||
environment:
|
||||
@ -100,13 +99,14 @@ services:
|
||||
- ADMIN_USERNAME=${DASHBOARD_ADMIN_USERNAME:-}
|
||||
- ADMIN_PASSWORD=${DASHBOARD_ADMIN_PASSWORD:-}
|
||||
- DATABASE_URL=mongodb://${MONGO_ADMIN_USERNAME}:${MONGO_ADMIN_PASSWORD}@mongo:27017/?replicaSet=rs0&readPreference=primaryPreferred
|
||||
volumes:
|
||||
- /etc/localtime:/etc/localtime:ro
|
||||
depends_on:
|
||||
setup:
|
||||
condition: service_completed_successfully
|
||||
|
||||
openvidu:
|
||||
image: docker.io/openvidu/openvidu-server:3.4.1
|
||||
platform: linux/amd64
|
||||
image: docker.io/openvidu/openvidu-server:3.5.0
|
||||
restart: unless-stopped
|
||||
container_name: openvidu
|
||||
extra_hosts:
|
||||
@ -123,13 +123,13 @@ services:
|
||||
volumes:
|
||||
- ./livekit.yaml:/etc/livekit.yaml
|
||||
- ./scripts/entrypoint_openvidu.sh:/scripts/entrypoint.sh
|
||||
- /etc/localtime:/etc/localtime:ro
|
||||
depends_on:
|
||||
setup:
|
||||
condition: service_completed_successfully
|
||||
|
||||
ingress:
|
||||
image: docker.io/openvidu/ingress:3.4.1
|
||||
platform: linux/amd64
|
||||
image: docker.io/openvidu/ingress:3.5.0
|
||||
container_name: ingress
|
||||
restart: unless-stopped
|
||||
extra_hosts:
|
||||
@ -142,13 +142,13 @@ services:
|
||||
- INGRESS_CONFIG_FILE=/etc/ingress.yaml
|
||||
volumes:
|
||||
- ./ingress.yaml:/etc/ingress.yaml
|
||||
- /etc/localtime:/etc/localtime:ro
|
||||
depends_on:
|
||||
setup:
|
||||
condition: service_completed_successfully
|
||||
|
||||
egress:
|
||||
image: docker.io/livekit/egress:v1.10.0
|
||||
platform: linux/amd64
|
||||
image: docker.io/openvidu/egress:3.5.0
|
||||
restart: unless-stopped
|
||||
container_name: egress
|
||||
extra_hosts:
|
||||
@ -158,13 +158,13 @@ services:
|
||||
volumes:
|
||||
- ./egress.yaml:/etc/egress.yaml
|
||||
- egress-data:/home/egress/tmp
|
||||
- /etc/localtime:/etc/localtime:ro
|
||||
depends_on:
|
||||
setup:
|
||||
condition: service_completed_successfully
|
||||
|
||||
openvidu-meet:
|
||||
image: docker.io/openvidu/openvidu-meet:3.4.1
|
||||
platform: linux/amd64
|
||||
image: docker.io/openvidu/openvidu-meet:3.5.0
|
||||
container_name: openvidu-meet
|
||||
restart: on-failure
|
||||
ports:
|
||||
@ -199,25 +199,26 @@ services:
|
||||
- MEET_REDIS_PORT=6379
|
||||
- MEET_REDIS_PASSWORD=${REDIS_PASSWORD:-}
|
||||
- MEET_REDIS_DB=0
|
||||
- MEET_MONGO_URI=mongodb://${MONGO_ADMIN_USERNAME}:${MONGO_ADMIN_PASSWORD}@mongo:27017/?replicaSet=rs0&readPreference=primaryPreferred
|
||||
volumes:
|
||||
- ./scripts/entrypoint_openvidu_meet.sh:/scripts/entrypoint.sh
|
||||
- ./scripts/utils.sh:/scripts/utils.sh
|
||||
- /etc/localtime:/etc/localtime:ro
|
||||
entrypoint: /bin/sh /scripts/entrypoint.sh
|
||||
depends_on:
|
||||
setup:
|
||||
condition: service_completed_successfully
|
||||
|
||||
operator:
|
||||
image: docker.io/openvidu/openvidu-operator:3.4.1
|
||||
platform: linux/amd64
|
||||
image: docker.io/openvidu/openvidu-operator:3.5.0
|
||||
container_name: operator
|
||||
restart: unless-stopped
|
||||
volumes:
|
||||
- /var/run/docker.sock:/var/run/docker.sock
|
||||
- agents-config:/agents-config
|
||||
- ./:/deployment
|
||||
- /etc/localtime:/etc/localtime:ro
|
||||
environment:
|
||||
- PLATFORM=linux/amd64
|
||||
- MODE=agent-manager-local
|
||||
- DEPLOYMENT_FILES_DIR=/deployment
|
||||
- AGENTS_CONFIG_DIR=/agents-config
|
||||
@ -233,10 +234,11 @@ services:
|
||||
condition: service_completed_successfully
|
||||
|
||||
ready-check:
|
||||
image: docker.io/openvidu/openvidu-operator:3.4.1
|
||||
platform: linux/amd64
|
||||
image: docker.io/openvidu/openvidu-operator:3.5.0
|
||||
container_name: ready-check
|
||||
restart: on-failure
|
||||
volumes:
|
||||
- /etc/localtime:/etc/localtime:ro
|
||||
environment:
|
||||
- MODE=local-ready-check
|
||||
- OPENVIDU_ENVIRONMENT=local-platform
|
||||
@ -263,7 +265,6 @@ services:
|
||||
|
||||
setup:
|
||||
image: docker.io/busybox:1.37.0
|
||||
platform: linux/amd64
|
||||
container_name: setup
|
||||
restart: "no"
|
||||
volumes:
|
||||
@ -271,6 +272,7 @@ services:
|
||||
- mongo-data:/mongo
|
||||
- egress-data:/egress
|
||||
- ./scripts/setup.sh:/scripts/setup.sh
|
||||
- /etc/localtime:/etc/localtime:ro
|
||||
environment:
|
||||
- USE_HTTPS=${USE_HTTPS:-false}
|
||||
- LAN_MODE=${LAN_MODE:-false}
|
||||
|
||||
@ -1,5 +1,5 @@
|
||||
# Docker image of the agent.
|
||||
docker_image: docker.io/openvidu/agent-speech-processing:3.4.1
|
||||
docker_image: docker.io/openvidu/agent-speech-processing:3.5.0
|
||||
|
||||
# Whether to run the agent or not.
|
||||
enabled: false
|
||||
@ -16,7 +16,7 @@ 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, openai, groq, deepgram, assemblyai, fal, clova, speechmatics, gladia, sarvam, spitch]
|
||||
# Which speech-to-text AI provider to use [aws, azure, google, openai, azure_openai, groq, deepgram, assemblyai, fal, clova, speechmatics, gladia, sarvam, mistralai, cartesia, soniox]
|
||||
# The custom configuration for the selected provider must be set below
|
||||
provider:
|
||||
|
||||
@ -63,6 +63,10 @@ live_captions:
|
||||
# Removes profanity (swearing), or replaces letters of profane words with stars. Valid values: Masked | Removed | Raw
|
||||
# See https://learn.microsoft.com/en-us/azure/ai-services/translator/profanity-filtering
|
||||
profanity:
|
||||
# List of words or phrases to boost recognition accuracy. Azure will give higher priority to these phrases during recognition.
|
||||
phrase_list:
|
||||
# Controls punctuation behavior. If True, enables explicit punctuation mode where punctuation marks are added explicitly. If False (default), uses Azure's default punctuation behavior.
|
||||
explicit_punctuation:
|
||||
|
||||
azure_openai:
|
||||
# Credentials for Azure OpenAI APIs. See https://learn.microsoft.com/en-us/azure/api-management/api-management-authenticate-authorize-azure-openai
|
||||
@ -82,6 +86,8 @@ live_captions:
|
||||
project:
|
||||
# The language code to use for transcription (e.g., "en" for English).
|
||||
language:
|
||||
# Whether to automatically detect the language.
|
||||
detect_language:
|
||||
# ID of the model to use for speech-to-text.
|
||||
model:
|
||||
# Initial prompt to guide the transcription.
|
||||
@ -135,6 +141,8 @@ live_captions:
|
||||
# The language of the input audio. Supplying the input language in ISO-639-1 format
|
||||
# (https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) will improve accuracy and latency.
|
||||
language:
|
||||
# Whether to automatically detect the language.
|
||||
detect_language:
|
||||
# Optional text prompt to guide the transcription. Only supported for whisper-1.
|
||||
prompt:
|
||||
|
||||
@ -146,8 +154,12 @@ live_captions:
|
||||
# The language of the input audio. Supplying the input language in ISO-639-1 format
|
||||
# (https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) will improve accuracy and latency.
|
||||
language:
|
||||
# Whether to automatically detect the language.
|
||||
detect_language:
|
||||
# Prompt to guide the model's style or specify how to spell unfamiliar words. 224 tokens max.
|
||||
prompt:
|
||||
# Base URL for the Groq API. By default "https://api.groq.com/openai/v1"
|
||||
base_url:
|
||||
|
||||
deepgram:
|
||||
# See https://console.deepgram.com/
|
||||
@ -156,25 +168,27 @@ live_captions:
|
||||
model:
|
||||
# See https://developers.deepgram.com/reference/speech-to-text-api/listen-streaming#request.query.language
|
||||
language:
|
||||
# Whether to enable automatic language detection. Defaults to false. See https://developers.deepgram.com/docs/language-detection
|
||||
# Whether to enable automatic language detection. See https://developers.deepgram.com/docs/language-detection
|
||||
detect_language: false
|
||||
# Whether to return interim (non-final) transcription results. Defaults to true. See https://developers.deepgram.com/docs/interim-results
|
||||
# Whether to return interim (non-final) transcription results. See https://developers.deepgram.com/docs/interim-results
|
||||
interim_results: true
|
||||
# Whether to apply smart formatting to numbers, dates, etc. Defaults to false. See https://developers.deepgram.com/docs/smart-format
|
||||
# Whether to apply smart formatting to numbers, dates, etc. 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
|
||||
# When smart_format is used, ensures it does not wait for sequence to be complete before returning results. 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
|
||||
# Whether to add punctuations to the transcription. Turn detector will work better with punctuations. See https://developers.deepgram.com/docs/punctuation
|
||||
punctuate: true
|
||||
# Whether to include filler words (um, uh, etc.) in transcription. Defaults to true. See https://developers.deepgram.com/docs/filler-words
|
||||
# Whether to include filler words (um, uh, etc.) in transcription. See https://developers.deepgram.com/docs/filler-words
|
||||
filler_words: true
|
||||
# Whether to filter profanity from the transcription. Defaults to false. See https://developers.deepgram.com/docs/profanity-filter
|
||||
# Whether to filter profanity from the transcription. See https://developers.deepgram.com/docs/profanity-filter
|
||||
profanity_filter: false
|
||||
# List of tuples containing keywords and their boost values for improved recognition. Each tuple should be (keyword: str, boost: float). Defaults to None. keywords does not work with Nova-3 models. Use keyterms instead.
|
||||
# Whether to transcribe numbers as numerals. See https://developers.deepgram.com/docs/numerals
|
||||
numerals: false
|
||||
# List of tuples containing keywords and their boost values for improved recognition. Each tuple should be (keyword: str, boost: float). keywords does not work with Nova-3 models. Use keyterms instead.
|
||||
# keywords:
|
||||
# - [OpenVidu, 1.5]
|
||||
# - [WebRTC, 1]
|
||||
# List of key terms to improve recognition accuracy. Defaults to None. keyterms is supported by Nova-3 models.
|
||||
# List of key terms to improve recognition accuracy. keyterms is supported by Nova-3 models.
|
||||
# Commented below is an example
|
||||
keyterms:
|
||||
# - "OpenVidu"
|
||||
@ -183,8 +197,18 @@ live_captions:
|
||||
assemblyai:
|
||||
# API key for AssemblyAI. See https://www.assemblyai.com/dashboard/api-keys
|
||||
api_key:
|
||||
# The confidence threshold (0.0 to 1.0) to use when determining if the end of a turn has been reached.
|
||||
end_of_turn_confidence_threshold:
|
||||
# The minimum amount of silence in milliseconds required to detect end of turn when confident.
|
||||
min_end_of_turn_silence_when_confident:
|
||||
# The maximum amount of silence in milliseconds allowed in a turn before end of turn is triggered.
|
||||
max_turn_silence:
|
||||
# Whether to return formatted final transcripts (proper punctuation, letter casing...). If enabled, formatted final transcripts are emitted shortly following an end-of-turn detection.
|
||||
format_turns: true
|
||||
# List of keyterms to improve recognition accuracy for specific words and phrases.
|
||||
keyterms_prompt:
|
||||
# - "OpenVidu"
|
||||
# - "WebRTC"
|
||||
|
||||
fal:
|
||||
# API key for fal. See https://fal.ai/dashboard/keys
|
||||
@ -208,12 +232,14 @@ live_captions:
|
||||
speechmatics:
|
||||
# API key for Speechmatics. See https://portal.speechmatics.com/manage-access/
|
||||
api_key:
|
||||
# ISO 639-1 language code. All languages are global and can understand different dialects/accents. To see the list of all supported languages, see https://docs.speechmatics.com/introduction/supported-languages
|
||||
# ISO 639-1 language code. All languages are global and can understand different dialects/accents. To see the list of all supported languages, see https://docs.speechmatics.com/speech-to-text/languages#transcription-languages
|
||||
language:
|
||||
# Operating point to use for the transcription per required accuracy & complexity. To learn more, see https://docs.speechmatics.com/features/accuracy-language-packs#accuracy
|
||||
# Operating point to use for the transcription per required accuracy & complexity. To learn more, see https://docs.speechmatics.com/speech-to-text/languages#operating-points
|
||||
operating_point:
|
||||
# Partial transcripts allow you to receive preliminary transcriptions and update as more context is available until the higher-accuracy final transcript is returned. Partials are returned faster but without any post-processing such as formatting. See https://docs.speechmatics.com/features/realtime-latency#partial-transcripts
|
||||
# Partial transcripts allow you to receive preliminary transcriptions and update as more context is available until the higher-accuracy final transcript is returned. Partials are returned faster but without any post-processing such as formatting. See https://docs.speechmatics.com/speech-to-text/realtime/output#partial-transcripts
|
||||
enable_partials:
|
||||
# Enable speaker diarization. When enabled, the STT engine will determine and attribute words to unique speakers. The speaker_sensitivity parameter can be used to adjust the sensitivity of diarization
|
||||
enable_diarization:
|
||||
# RFC-5646 language code to make spelling rules more consistent in the transcription output. See https://docs.speechmatics.com/features/word-tagging#output-locale
|
||||
output_locale:
|
||||
# The delay in seconds between the end of a spoken word and returning the final transcript results. See https://docs.speechmatics.com/features/realtime-latency#configuration-example
|
||||
@ -255,6 +281,10 @@ live_captions:
|
||||
languages:
|
||||
# Whether to allow switching between languages during recognition. Defaults to True
|
||||
code_switching:
|
||||
# https://docs.gladia.io/api-reference/v2/live/init#body-pre-processing-audio-enhancer
|
||||
pre_processing_audio_enhancer:
|
||||
# https://docs.gladia.io/api-reference/v2/live/init#body-pre-processing-speech-threshold
|
||||
pre_processing_speech_threshold:
|
||||
|
||||
sarvam:
|
||||
# API key for Sarvam. See https://dashboard.sarvam.ai/key-management
|
||||
@ -263,3 +293,29 @@ 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:
|
||||
|
||||
mistralai:
|
||||
# API key for Mistral AI. See https://console.mistral.ai/api-keys
|
||||
api_key:
|
||||
# Name of the Voxtral STT model to use. Default to voxtral-mini-latest. See https://docs.mistral.ai/capabilities/audio/
|
||||
model:
|
||||
# The language code to use for transcription (e.g., "en" for English)
|
||||
language:
|
||||
|
||||
cartesia:
|
||||
# API key for Cartesia. See https://play.cartesia.ai/keys
|
||||
api_key:
|
||||
# The Cartesia STT model to use
|
||||
model:
|
||||
# The language code to use for transcription (e.g., "en" for English)
|
||||
language:
|
||||
|
||||
soniox:
|
||||
# API key for Soniox. See https://console.soniox.com/
|
||||
api_key:
|
||||
# Set language hints when possible to significantly improve accuracy. See: https://soniox.com/docs/stt/concepts/language-hints
|
||||
language_hints:
|
||||
# - "en"
|
||||
# - "es"
|
||||
# Set context to improve recognition of difficult and rare words. Context is a string and can include words, phrases, sentences, or summaries (limit: 10K chars). See https://soniox.com/docs/stt/concepts/context
|
||||
context:
|
||||
@ -14,14 +14,4 @@ if [ -z "$LAN_PRIVATE_IP" ]; then
|
||||
fi
|
||||
|
||||
# Replace the LAN_PRIVATE_IP in the .env file
|
||||
sed -i'' -e "s/LAN_PRIVATE_IP=.*/LAN_PRIVATE_IP=$LAN_PRIVATE_IP/g" .env
|
||||
|
||||
# If sillicon mac, enable EXPERIMENTAL_DOCKER_DESKTOP_FORCE_QEMU flag
|
||||
if [ "$(uname -m)" = "arm64" ]; then
|
||||
if ! grep -q "EXPERIMENTAL_DOCKER_DESKTOP_FORCE_QEMU" .env; then
|
||||
echo "# Enable this flag to run Docker Desktop on Apple Silicon Macs" >> .env
|
||||
echo "EXPERIMENTAL_DOCKER_DESKTOP_FORCE_QEMU=1" >> .env
|
||||
else
|
||||
sed -i'' -e "s/EXPERIMENTAL_DOCKER_DESKTOP_FORCE_QEMU=.*/EXPERIMENTAL_DOCKER_DESKTOP_FORCE_QEMU=1/g" .env
|
||||
fi
|
||||
fi
|
||||
sed -i'' -e "s/LAN_PRIVATE_IP=.*/LAN_PRIVATE_IP=$LAN_PRIVATE_IP/g" .env
|
||||
@ -1,7 +1,6 @@
|
||||
services:
|
||||
caddy-proxy:
|
||||
image: docker.io/openvidu/openvidu-caddy-local:3.4.1
|
||||
platform: linux/amd64
|
||||
image: docker.io/openvidu/openvidu-caddy-local:3.5.0
|
||||
container_name: caddy-proxy
|
||||
restart: unless-stopped
|
||||
extra_hosts:
|
||||
@ -23,6 +22,7 @@ services:
|
||||
- MEET_INITIAL_API_KEY=${MEET_INITIAL_API_KEY:-}
|
||||
volumes:
|
||||
- ./custom-layout:/var/www/custom-layout
|
||||
- /etc/localtime:/etc/localtime:ro
|
||||
ports:
|
||||
- 5443:5443
|
||||
- 6443:6443
|
||||
@ -34,14 +34,14 @@ services:
|
||||
condition: service_completed_successfully
|
||||
|
||||
redis:
|
||||
image: docker.io/redis:7.4.4-alpine
|
||||
platform: linux/amd64
|
||||
image: docker.io/redis:8.2.2-alpine
|
||||
container_name: redis
|
||||
restart: unless-stopped
|
||||
ports:
|
||||
- 6379:6379
|
||||
volumes:
|
||||
- redis:/data
|
||||
- /etc/localtime:/etc/localtime:ro
|
||||
command: >
|
||||
redis-server
|
||||
--bind 0.0.0.0
|
||||
@ -51,8 +51,7 @@ services:
|
||||
condition: service_completed_successfully
|
||||
|
||||
minio:
|
||||
image: docker.io/openvidu/minio:2025.5.24-debian-12-r1
|
||||
platform: linux/amd64
|
||||
image: docker.io/openvidu/minio:2025.9.7-debian-12-r3
|
||||
restart: unless-stopped
|
||||
ports:
|
||||
- 9000:9000
|
||||
@ -61,23 +60,25 @@ services:
|
||||
- MINIO_ROOT_PASSWORD=${MINIO_SECRET_KEY:-}
|
||||
- MINIO_DEFAULT_BUCKETS=openvidu-appdata
|
||||
- MINIO_CONSOLE_SUBPATH=/minio-console
|
||||
- MINIO_BROWSER=on
|
||||
- MINIO_BROWSER_REDIRECT_URL=http://localhost:7880/minio-console
|
||||
volumes:
|
||||
- minio-data:/bitnami/minio/data
|
||||
- minio-certs:/certs
|
||||
- /etc/localtime:/etc/localtime:ro
|
||||
depends_on:
|
||||
setup:
|
||||
condition: service_completed_successfully
|
||||
|
||||
mongo:
|
||||
image: docker.io/openvidu/mongodb:8.0.12-r0
|
||||
platform: linux/amd64
|
||||
image: docker.io/openvidu/mongodb:8.0.15-r0
|
||||
container_name: mongo
|
||||
restart: unless-stopped
|
||||
ports:
|
||||
- 27017:27017
|
||||
volumes:
|
||||
- mongo-data:/bitnami/mongodb
|
||||
- /etc/localtime:/etc/localtime:ro
|
||||
environment:
|
||||
- MONGODB_ROOT_USER=${MONGO_ADMIN_USERNAME:-}
|
||||
- MONGODB_ROOT_PASSWORD=${MONGO_ADMIN_PASSWORD:-}
|
||||
@ -85,14 +86,12 @@ services:
|
||||
- MONGODB_REPLICA_SET_MODE=primary
|
||||
- MONGODB_REPLICA_SET_NAME=rs0
|
||||
- MONGODB_REPLICA_SET_KEY=devreplicasetkey
|
||||
- EXPERIMENTAL_DOCKER_DESKTOP_FORCE_QEMU=${EXPERIMENTAL_DOCKER_DESKTOP_FORCE_QEMU:-0}
|
||||
depends_on:
|
||||
setup:
|
||||
condition: service_completed_successfully
|
||||
|
||||
dashboard:
|
||||
image: docker.io/openvidu/openvidu-dashboard:3.4.1
|
||||
platform: linux/amd64
|
||||
image: docker.io/openvidu/openvidu-dashboard:3.5.0
|
||||
container_name: dashboard
|
||||
restart: unless-stopped
|
||||
environment:
|
||||
@ -100,13 +99,14 @@ services:
|
||||
- ADMIN_USERNAME=${DASHBOARD_ADMIN_USERNAME:-}
|
||||
- ADMIN_PASSWORD=${DASHBOARD_ADMIN_PASSWORD:-}
|
||||
- DATABASE_URL=mongodb://${MONGO_ADMIN_USERNAME}:${MONGO_ADMIN_PASSWORD}@mongo:27017/?replicaSet=rs0&readPreference=primaryPreferred
|
||||
volumes:
|
||||
- /etc/localtime:/etc/localtime:ro
|
||||
depends_on:
|
||||
setup:
|
||||
condition: service_completed_successfully
|
||||
|
||||
openvidu:
|
||||
image: docker.io/openvidu/openvidu-server-pro:3.4.1
|
||||
platform: linux/amd64
|
||||
image: docker.io/openvidu/openvidu-server-pro:3.5.0
|
||||
restart: unless-stopped
|
||||
container_name: openvidu
|
||||
extra_hosts:
|
||||
@ -125,13 +125,13 @@ services:
|
||||
volumes:
|
||||
- ./livekit.yaml:/etc/livekit.yaml
|
||||
- ./scripts/entrypoint_openvidu.sh:/scripts/entrypoint.sh
|
||||
- /etc/localtime:/etc/localtime:ro
|
||||
depends_on:
|
||||
setup:
|
||||
condition: service_completed_successfully
|
||||
|
||||
ingress:
|
||||
image: docker.io/openvidu/ingress:3.4.1
|
||||
platform: linux/amd64
|
||||
image: docker.io/openvidu/ingress:3.5.0
|
||||
container_name: ingress
|
||||
restart: unless-stopped
|
||||
extra_hosts:
|
||||
@ -144,13 +144,13 @@ services:
|
||||
- INGRESS_CONFIG_FILE=/etc/ingress.yaml
|
||||
volumes:
|
||||
- ./ingress.yaml:/etc/ingress.yaml
|
||||
- /etc/localtime:/etc/localtime:ro
|
||||
depends_on:
|
||||
setup:
|
||||
condition: service_completed_successfully
|
||||
|
||||
egress:
|
||||
image: docker.io/livekit/egress:v1.10.0
|
||||
platform: linux/amd64
|
||||
image: docker.io/openvidu/egress:3.5.0
|
||||
restart: unless-stopped
|
||||
container_name: egress
|
||||
extra_hosts:
|
||||
@ -160,13 +160,13 @@ services:
|
||||
volumes:
|
||||
- ./egress.yaml:/etc/egress.yaml
|
||||
- egress-data:/home/egress
|
||||
- /etc/localtime:/etc/localtime:ro
|
||||
depends_on:
|
||||
setup:
|
||||
condition: service_completed_successfully
|
||||
|
||||
openvidu-meet:
|
||||
image: docker.io/openvidu/openvidu-meet:3.4.1
|
||||
platform: linux/amd64
|
||||
image: docker.io/openvidu/openvidu-meet:3.5.0
|
||||
container_name: openvidu-meet
|
||||
restart: on-failure
|
||||
ports:
|
||||
@ -201,17 +201,18 @@ services:
|
||||
- MEET_REDIS_PORT=6379
|
||||
- MEET_REDIS_PASSWORD=${REDIS_PASSWORD:-}
|
||||
- MEET_REDIS_DB=0
|
||||
- MEET_MONGO_URI=mongodb://${MONGO_ADMIN_USERNAME}:${MONGO_ADMIN_PASSWORD}@mongo:27017/?replicaSet=rs0&readPreference=primaryPreferred
|
||||
volumes:
|
||||
- ./scripts/entrypoint_openvidu_meet.sh:/scripts/entrypoint.sh
|
||||
- ./scripts/utils.sh:/scripts/utils.sh
|
||||
- /etc/localtime:/etc/localtime:ro
|
||||
entrypoint: /bin/sh /scripts/entrypoint.sh
|
||||
depends_on:
|
||||
setup:
|
||||
condition: service_completed_successfully
|
||||
|
||||
openvidu-v2compatibility:
|
||||
image: docker.io/openvidu/openvidu-v2compatibility:3.4.1
|
||||
platform: linux/amd64
|
||||
image: docker.io/openvidu/openvidu-v2compatibility:3.5.0
|
||||
restart: unless-stopped
|
||||
container_name: openvidu-v2compatibility
|
||||
entrypoint: /bin/sh /scripts/entrypoint.sh
|
||||
@ -247,15 +248,17 @@ services:
|
||||
- ./recordings:/opt/openvidu/recordings
|
||||
- ./scripts/entrypoint_v2comp.sh:/scripts/entrypoint.sh
|
||||
- ./scripts/utils.sh:/scripts/utils.sh
|
||||
- /etc/localtime:/etc/localtime:ro
|
||||
depends_on:
|
||||
setup:
|
||||
condition: service_completed_successfully
|
||||
|
||||
ready-check:
|
||||
image: docker.io/openvidu/openvidu-operator:3.4.1
|
||||
platform: linux/amd64
|
||||
image: docker.io/openvidu/openvidu-operator:3.5.0
|
||||
container_name: ready-check
|
||||
restart: on-failure
|
||||
volumes:
|
||||
- /etc/localtime:/etc/localtime:ro
|
||||
environment:
|
||||
- MODE=local-ready-check
|
||||
- OPENVIDU_ENVIRONMENT=local-platform
|
||||
@ -282,16 +285,15 @@ services:
|
||||
- mongo
|
||||
|
||||
operator:
|
||||
image: docker.io/openvidu/openvidu-operator:3.4.1
|
||||
platform: linux/amd64
|
||||
image: docker.io/openvidu/openvidu-operator:3.5.0
|
||||
container_name: operator
|
||||
restart: unless-stopped
|
||||
volumes:
|
||||
- /var/run/docker.sock:/var/run/docker.sock
|
||||
- agents-config:/agents-config
|
||||
- ./:/deployment
|
||||
- /etc/localtime:/etc/localtime:ro
|
||||
environment:
|
||||
- PLATFORM=linux/amd64
|
||||
- MODE=agent-manager-local
|
||||
- DEPLOYMENT_FILES_DIR=/deployment
|
||||
- AGENTS_CONFIG_DIR=/agents-config
|
||||
@ -307,7 +309,6 @@ services:
|
||||
condition: service_completed_successfully
|
||||
setup:
|
||||
image: docker.io/busybox:1.37.0
|
||||
platform: linux/amd64
|
||||
container_name: setup
|
||||
restart: "no"
|
||||
volumes:
|
||||
@ -315,6 +316,7 @@ services:
|
||||
- mongo-data:/mongo
|
||||
- egress-data:/egress
|
||||
- ./scripts/setup.sh:/scripts/setup.sh
|
||||
- /etc/localtime:/etc/localtime:ro
|
||||
environment:
|
||||
- USE_HTTPS=${USE_HTTPS:-false}
|
||||
- LAN_MODE=${LAN_MODE:-false}
|
||||
|
||||
Loading…
x
Reference in New Issue
Block a user