From a4c06a65020329fbc010bca0874ba905b90a5672 Mon Sep 17 00:00:00 2001 From: Georgios Hadjiharalambous Date: Fri, 11 Sep 2026 11:53:06 +0100 Subject: [PATCH 01/10] update dates --- docs/private/melia-1-realtime.mdx | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/docs/private/melia-1-realtime.mdx b/docs/private/melia-1-realtime.mdx index c4cbe21b..e2147ac7 100644 --- a/docs/private/melia-1-realtime.mdx +++ b/docs/private/melia-1-realtime.mdx @@ -127,8 +127,8 @@ Message format: | Word timings | ✅ | | | Language labeling | ✅ Word-level | | | Alphanumeric smart formatting | ✅ Improved accuracy, e.g. recognition of email/web addresses transcription | | -| Language hints (guide predicted languages) | ⚠️ Not yet | Targeting August 2026 | -| End of Utterance / Force end of Utterance | ⚠️ Not yet | Targeting August 2026 | +| Language hints (guide predicted languages) | ⚠️ Not yet | Targeting September 2026 | +| End of Utterance / Force end of Utterance | ⚠️ Not yet | Targeting September 2026 | | Speaker diarization | ⚠️ Not yet | Planned Q4 2026 | | Speaker identification | ⚠️ Not yet | Prioritized according to customer need | | `max_delay` / `max_delay_mode` latency control | ⚠️ Not yet. Finals latency averages around 4 seconds and varies | Lower latency and control planned Q3/Q4 2026 | From 924b813ad19fa4c0e1ed52d598055489b85f4f2f Mon Sep 17 00:00:00 2001 From: Georgios Hadjiharalambous Date: Fri, 11 Sep 2026 13:50:06 +0100 Subject: [PATCH 02/10] add melia rt on prem docs --- .../container/accessing-images.mdx | 2 +- .../container/gpu-speech-to-text-melia-1.mdx | 63 +++++++++++++++++-- 2 files changed, 60 insertions(+), 5 deletions(-) diff --git a/docs/deployments/container/accessing-images.mdx b/docs/deployments/container/accessing-images.mdx index fda15cae..6246d102 100644 --- a/docs/deployments/container/accessing-images.mdx +++ b/docs/deployments/container/accessing-images.mdx @@ -131,7 +131,7 @@ This image contains both models. To load only one of them, see [Running only one ### Melia 1 model -Melia 1 is a multilingual, GPU-only model, currently available in Batch. For an overview, see [Models](/speech-to-text/models#melia-1). +Melia 1 is a multilingual, GPU-only model, generally available in Batch, with Real-time (streaming) available as an early preview. For an overview, see [Models](/speech-to-text/models#melia-1). {`# pulling the Melia 1 Batch transcriber with the ${smVariables.latestMelia1ContainerVersion} tag: diff --git a/docs/deployments/container/gpu-speech-to-text-melia-1.mdx b/docs/deployments/container/gpu-speech-to-text-melia-1.mdx index 66741298..90f524e4 100644 --- a/docs/deployments/container/gpu-speech-to-text-melia-1.mdx +++ b/docs/deployments/container/gpu-speech-to-text-melia-1.mdx @@ -10,7 +10,7 @@ import melia1GpuSttDockerCompose from './assets/gpu-stt-melia-1-docker-compose.y # GPU Speech to text container (Melia 1) -Melia 1 is a multilingual, GPU-only model, currently available in Batch. For an overview of the model, see [Models](/speech-to-text/models#melia-1). +Melia 1 is a multilingual, GPU-only model, generally available in Batch. For an overview of the model, see [Models](/speech-to-text/models#melia-1). Real-time (streaming) inference is available as an early preview — see [Real-time inference](#real-time-inference). ## Prerequisites @@ -85,20 +85,75 @@ I0705 08:12:55.598672 1 http_server.cc:400] "Started Metrics Service at 0.0.0.0: ### Batch inference -The Melia 1 inference server currently runs in _batch_ mode only, processing whole files and returning the transcript at the end. +The Melia 1 inference server processes whole files and returns the transcript at the end. For streaming, see [Real-time inference](#real-time-inference) below. -Once the inference server is running, run the Melia 1 transcriber with the `SM_INFERENCE_ENDPOINT` environment variable set to the gRPC endpoint of the inference server. The transcriber offloads inference to the server, so it does not require a GPU: +Once the inference server is running, run the Melia 1 transcriber with the `SM_INFERENCE_ENDPOINT` environment variable set to the gRPC endpoint of the inference server. The transcriber offloads inference to the server, so it does not require a GPU. The transcriber is a single unified image for both batch and real-time, so the `SM_ASR_MODE` environment variable must also be set, to `batch` or `rt`: ```bash docker run --rm -it \ -e SM_INFERENCE_ENDPOINT=: \ + -e SM_ASR_MODE=batch \ -v $PWD/license.json:/license.json \ -v $PWD/example.wav:/input.audio \ - + speechmaticspublic.azurecr.io/sm-asr-transcriber-melia-1:${smVariables.latestMelia1ContainerVersion} ``` To accept multiple jobs over an HTTP API without restarting the container between jobs, run the transcriber as a [batch persistent worker](/deployments/container/batch-persistent-worker). +### Real-time inference + +:::warning +Real-time (streaming) inference for Melia 1 is an early preview for evaluation only, and is not intended for production use. Feature coverage matches the [Melia 1 Realtime Preview](/private/melia-1-realtime). +::: + +The Melia 1 inference server can run in two modes: _batch_ or _real-time_. The default mode is batch. To configure the server for real-time, set the `SM_STREAMING` environment variable to `true` when starting the inference server: + + + {`docker run --rm -it \\ + -v $PWD/license.json:/license.json \\ + --gpus '"device=0"' \\ + -e CUDA_VISIBLE_DEVICES \\ + -e SM_STREAMING=true \\ + -p 8001:8001 \\ + speechmaticspublic.azurecr.io/sm-asr-inference-server-melia-1:${smVariables.latestMelia1ContainerVersion} +`} + + +When the Container starts you should see output similar to this, +indicating that the server has started and is ready to serve requests. + +``` +I0911 11:04:02.321449 1 server.cc:709] ++--------------------+---------+--------+ +| Model | Version | Status | ++--------------------+---------+--------+ +| aed | 1 | READY | +| body | 1 | READY | +| preprocessor | 1 | READY | +| streaming | 1 | READY | +| streaming_ensemble | 1 | READY | ++--------------------+---------+--------+ +... +I0911 11:04:02.449550 1 grpc_server.cc:2579] "Started GRPCInferenceService at 0.0.0.0:8001" +I0911 11:04:02.449744 1 http_server.cc:4961] "Started HTTPService at 0.0.0.0:8000" +I0911 11:04:02.491582 1 http_server.cc:400] "Started Metrics Service at 0.0.0.0:8002" +``` + +The server can only support one of these modes at once. + +Once the inference server is running in real-time mode, run the Melia 1 transcriber as normal, with the `SM_INFERENCE_ENDPOINT` environment variable set to the gRPC endpoint of the inference server. Unlike the [Standard and Enhanced GPU container](/deployments/container/gpu-speech-to-text#batch-and-realtime-inference), Melia 1 uses the same transcriber image for both batch and real-time — there are no separate `rt-` and `batch-` prefixed images. Set `SM_ASR_MODE=rt` to run the transcriber in real-time mode: + +```bash +docker run --rm -it \ + -e SM_INFERENCE_ENDPOINT=: \ + -e SM_ASR_MODE=rt \ + -v $PWD/license.json:/license.json \ + -p 9000:9000 \ + speechmaticspublic.azurecr.io/sm-asr-transcriber-melia-1:${smVariables.latestMelia1ContainerVersion} +``` + +Connect over WebSocket as described in [Realtime transcription](/deployments/container/cpu-speech-to-text#realtime-transcription). + ### Monitoring the server The inference server is based on [Nvidia's Triton architecture](https://developer.nvidia.com/nvidia-triton-inference-server) and as such From 96c3deebe99bc9726d511ecd3cbab96e2af9897f Mon Sep 17 00:00:00 2001 From: Georgios Hadjiharalambous Date: Fri, 11 Sep 2026 18:22:02 +0100 Subject: [PATCH 03/10] add multi session capability --- docs/deployments/container/gpu-speech-to-text-melia-1.mdx | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/docs/deployments/container/gpu-speech-to-text-melia-1.mdx b/docs/deployments/container/gpu-speech-to-text-melia-1.mdx index 90f524e4..2d7cd929 100644 --- a/docs/deployments/container/gpu-speech-to-text-melia-1.mdx +++ b/docs/deployments/container/gpu-speech-to-text-melia-1.mdx @@ -154,6 +154,12 @@ docker run --rm -it \ Connect over WebSocket as described in [Realtime transcription](/deployments/container/cpu-speech-to-text#realtime-transcription). +### Multi-session containers + +By default the real-time container will accept only one websocket connection at a time. To enable multiple connections, set the environment variable `SM_MAX_CONCURRENT_CONNECTIONS` to the maximum number of sessions to allow. When this is set, the `/ready` [health check endpoint](#health-service) will return true if there is a free connection available. + +CPU usage scales with the number of active sessions, whereas most memory usage is shared between connections. + ### Monitoring the server The inference server is based on [Nvidia's Triton architecture](https://developer.nvidia.com/nvidia-triton-inference-server) and as such From e96f807d4fc8a040e21604f4ea5f884ad8ced3b0 Mon Sep 17 00:00:00 2001 From: Georgios Hadjiharalambous Date: Fri, 11 Sep 2026 18:31:11 +0100 Subject: [PATCH 04/10] move to private area --- .../container/gpu-speech-to-text-melia-1.mdx | 64 +--------------- docs/private/melia-1-realtime.mdx | 76 +++++++++++++++++++ 2 files changed, 77 insertions(+), 63 deletions(-) diff --git a/docs/deployments/container/gpu-speech-to-text-melia-1.mdx b/docs/deployments/container/gpu-speech-to-text-melia-1.mdx index 2d7cd929..b2c9f5cd 100644 --- a/docs/deployments/container/gpu-speech-to-text-melia-1.mdx +++ b/docs/deployments/container/gpu-speech-to-text-melia-1.mdx @@ -98,69 +98,7 @@ docker run --rm -it \ speechmaticspublic.azurecr.io/sm-asr-transcriber-melia-1:${smVariables.latestMelia1ContainerVersion} ``` -To accept multiple jobs over an HTTP API without restarting the container between jobs, run the transcriber as a [batch persistent worker](/deployments/container/batch-persistent-worker). - -### Real-time inference - -:::warning -Real-time (streaming) inference for Melia 1 is an early preview for evaluation only, and is not intended for production use. Feature coverage matches the [Melia 1 Realtime Preview](/private/melia-1-realtime). -::: - -The Melia 1 inference server can run in two modes: _batch_ or _real-time_. The default mode is batch. To configure the server for real-time, set the `SM_STREAMING` environment variable to `true` when starting the inference server: - - - {`docker run --rm -it \\ - -v $PWD/license.json:/license.json \\ - --gpus '"device=0"' \\ - -e CUDA_VISIBLE_DEVICES \\ - -e SM_STREAMING=true \\ - -p 8001:8001 \\ - speechmaticspublic.azurecr.io/sm-asr-inference-server-melia-1:${smVariables.latestMelia1ContainerVersion} -`} - - -When the Container starts you should see output similar to this, -indicating that the server has started and is ready to serve requests. - -``` -I0911 11:04:02.321449 1 server.cc:709] -+--------------------+---------+--------+ -| Model | Version | Status | -+--------------------+---------+--------+ -| aed | 1 | READY | -| body | 1 | READY | -| preprocessor | 1 | READY | -| streaming | 1 | READY | -| streaming_ensemble | 1 | READY | -+--------------------+---------+--------+ -... -I0911 11:04:02.449550 1 grpc_server.cc:2579] "Started GRPCInferenceService at 0.0.0.0:8001" -I0911 11:04:02.449744 1 http_server.cc:4961] "Started HTTPService at 0.0.0.0:8000" -I0911 11:04:02.491582 1 http_server.cc:400] "Started Metrics Service at 0.0.0.0:8002" -``` - -The server can only support one of these modes at once. - -Once the inference server is running in real-time mode, run the Melia 1 transcriber as normal, with the `SM_INFERENCE_ENDPOINT` environment variable set to the gRPC endpoint of the inference server. Unlike the [Standard and Enhanced GPU container](/deployments/container/gpu-speech-to-text#batch-and-realtime-inference), Melia 1 uses the same transcriber image for both batch and real-time — there are no separate `rt-` and `batch-` prefixed images. Set `SM_ASR_MODE=rt` to run the transcriber in real-time mode: - -```bash -docker run --rm -it \ - -e SM_INFERENCE_ENDPOINT=: \ - -e SM_ASR_MODE=rt \ - -v $PWD/license.json:/license.json \ - -p 9000:9000 \ - speechmaticspublic.azurecr.io/sm-asr-transcriber-melia-1:${smVariables.latestMelia1ContainerVersion} -``` - -Connect over WebSocket as described in [Realtime transcription](/deployments/container/cpu-speech-to-text#realtime-transcription). - -### Multi-session containers - -By default the real-time container will accept only one websocket connection at a time. To enable multiple connections, set the environment variable `SM_MAX_CONCURRENT_CONNECTIONS` to the maximum number of sessions to allow. When this is set, the `/ready` [health check endpoint](#health-service) will return true if there is a free connection available. - -CPU usage scales with the number of active sessions, whereas most memory usage is shared between connections. - -### Monitoring the server +To accept multiple jobs over an HTTP API without restarting the container between jobs, run the transcriber as a [batch persistent worker](/deployments/container/batch-persistent-worker).### Monitoring the server The inference server is based on [Nvidia's Triton architecture](https://developer.nvidia.com/nvidia-triton-inference-server) and as such can be monitored using Triton's inbuilt Prometheus metrics, or the GRPC/HTTP APIs. To expose these, configure an external mapping for port diff --git a/docs/private/melia-1-realtime.mdx b/docs/private/melia-1-realtime.mdx index e2147ac7..2920c369 100644 --- a/docs/private/melia-1-realtime.mdx +++ b/docs/private/melia-1-realtime.mdx @@ -35,6 +35,82 @@ The preview runs on a dedicated endpoint, separate from the production Realtime The preview is deployed in France (EU) and Oregon (USA), and is available on SaaS only. +### Real-time inference on prem + +:::warning +Real-time (streaming) inference on prem for Melia 1 is an early preview for evaluation only, and is not intended for production use. +::: + +The Melia 1 inference server can run in two modes: _batch_ or _real-time_. The default mode is batch. To configure the server for real-time, set the `SM_STREAMING` environment variable to `true` when starting the inference server: + + + {`docker run --rm -it \\ + -v $PWD/license.json:/license.json \\ + --gpus '"device=0"' \\ + -e CUDA_VISIBLE_DEVICES \\ + -e SM_STREAMING=true \\ + -p 8001:8001 \\ + speechmaticspublic.azurecr.io/sm-asr-inference-server-melia-1:${smVariables.latestMelia1ContainerVersion} +`} + + +When the Container starts you should see output similar to this, +indicating that the server has started and is ready to serve requests. + +``` +I0911 11:04:02.321449 1 server.cc:709] ++--------------------+---------+--------+ +| Model | Version | Status | ++--------------------+---------+--------+ +| aed | 1 | READY | +| body | 1 | READY | +| preprocessor | 1 | READY | +| streaming | 1 | READY | +| streaming_ensemble | 1 | READY | ++--------------------+---------+--------+ +... +I0911 11:04:02.449550 1 grpc_server.cc:2579] "Started GRPCInferenceService at 0.0.0.0:8001" +I0911 11:04:02.449744 1 http_server.cc:4961] "Started HTTPService at 0.0.0.0:8000" +I0911 11:04:02.491582 1 http_server.cc:400] "Started Metrics Service at 0.0.0.0:8002" +``` + +The server can only support one of these modes at once. + +Once the inference server is running in real-time mode, run the Melia 1 transcriber as normal, with the `SM_INFERENCE_ENDPOINT` environment variable set to the gRPC endpoint of the inference server. Unlike the [Standard and Enhanced GPU container](/deployments/container/gpu-speech-to-text#batch-and-realtime-inference), Melia 1 uses the same transcriber image for both batch and real-time — there are no separate `rt-` and `batch-` prefixed images. Set `SM_ASR_MODE=rt` to run the transcriber in real-time mode: + +```bash +docker run --rm -it \ + -e SM_INFERENCE_ENDPOINT=: \ + -e SM_ASR_MODE=rt \ + -v $PWD/license.json:/license.json \ + -p 9000:9000 \ + speechmaticspublic.azurecr.io/sm-asr-transcriber-melia-1:${smVariables.latestMelia1ContainerVersion} +``` + +Connect over WebSocket as described in [Realtime transcription](/deployments/container/cpu-speech-to-text#realtime-transcription). + +### Multi-session containers + +By default the real-time container will accept only one websocket connection at a time. To enable multiple connections, set the environment variable `SM_MAX_CONCURRENT_CONNECTIONS` to the maximum number of sessions to allow. When this is set, the `/ready` [health check endpoint](#health-service) will return true if there is a free connection available. + +CPU usage scales with the number of active sessions, whereas most memory usage is shared between connections. + +### Monitoring the server + +The inference server is based on [Nvidia's Triton architecture](https://developer.nvidia.com/nvidia-triton-inference-server) and as such +can be monitored using Triton's inbuilt Prometheus metrics, or the GRPC/HTTP APIs. To expose these, configure an external mapping for port +8002(Prometheus) or 8000(HTTP). + +## Docker compose example + +This Docker Compose file will create a Speechmatics Melia 1 GPU Inference Server: + +(assumes your `license.json` file is in the current working directory) + + +{melia1GpuSttDockerCompose.replace('{{ CONTAINER_VERSION }}', smVariables.latestMelia1ContainerVersion)} + + ### Minimal example Example minimal `StartRecognition` message: From ef5c60e23451c3526182beb164132ac73fad6516 Mon Sep 17 00:00:00 2001 From: Georgios Hadjiharalambous Date: Fri, 11 Sep 2026 18:34:12 +0100 Subject: [PATCH 05/10] small improvement as public was referencing rt melia --- docs/deployments/container/gpu-speech-to-text-melia-1.mdx | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/deployments/container/gpu-speech-to-text-melia-1.mdx b/docs/deployments/container/gpu-speech-to-text-melia-1.mdx index b2c9f5cd..fe022ca5 100644 --- a/docs/deployments/container/gpu-speech-to-text-melia-1.mdx +++ b/docs/deployments/container/gpu-speech-to-text-melia-1.mdx @@ -85,7 +85,7 @@ I0705 08:12:55.598672 1 http_server.cc:400] "Started Metrics Service at 0.0.0.0: ### Batch inference -The Melia 1 inference server processes whole files and returns the transcript at the end. For streaming, see [Real-time inference](#real-time-inference) below. +The Melia 1 inference server currently runs in _batch_ mode only, processing whole files and returning the transcript at the end. Once the inference server is running, run the Melia 1 transcriber with the `SM_INFERENCE_ENDPOINT` environment variable set to the gRPC endpoint of the inference server. The transcriber offloads inference to the server, so it does not require a GPU. The transcriber is a single unified image for both batch and real-time, so the `SM_ASR_MODE` environment variable must also be set, to `batch` or `rt`: From 33b6a9fc2511cf4f4d7a63a6425c7d7d80b55843 Mon Sep 17 00:00:00 2001 From: Georgios Hadjiharalambous Date: Fri, 11 Sep 2026 18:35:04 +0100 Subject: [PATCH 06/10] formatting fix --- docs/deployments/container/gpu-speech-to-text-melia-1.mdx | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/docs/deployments/container/gpu-speech-to-text-melia-1.mdx b/docs/deployments/container/gpu-speech-to-text-melia-1.mdx index fe022ca5..c77e759e 100644 --- a/docs/deployments/container/gpu-speech-to-text-melia-1.mdx +++ b/docs/deployments/container/gpu-speech-to-text-melia-1.mdx @@ -98,7 +98,9 @@ docker run --rm -it \ speechmaticspublic.azurecr.io/sm-asr-transcriber-melia-1:${smVariables.latestMelia1ContainerVersion} ``` -To accept multiple jobs over an HTTP API without restarting the container between jobs, run the transcriber as a [batch persistent worker](/deployments/container/batch-persistent-worker).### Monitoring the server +To accept multiple jobs over an HTTP API without restarting the container between jobs, run the transcriber as a [batch persistent worker](/deployments/container/batch-persistent-worker). + +### Monitoring the server The inference server is based on [Nvidia's Triton architecture](https://developer.nvidia.com/nvidia-triton-inference-server) and as such can be monitored using Triton's inbuilt Prometheus metrics, or the GRPC/HTTP APIs. To expose these, configure an external mapping for port From 7123d5d19f6104d6e6077c964b133bd8d42af26b Mon Sep 17 00:00:00 2001 From: Georgios Hadjiharalambous Date: Fri, 11 Sep 2026 19:19:42 +0100 Subject: [PATCH 07/10] fixes --- docs/deployments/container/gpu-speech-to-text-melia-1.mdx | 3 +-- docs/private/melia-1-realtime.mdx | 4 +++- 2 files changed, 4 insertions(+), 3 deletions(-) diff --git a/docs/deployments/container/gpu-speech-to-text-melia-1.mdx b/docs/deployments/container/gpu-speech-to-text-melia-1.mdx index c77e759e..d3d125e1 100644 --- a/docs/deployments/container/gpu-speech-to-text-melia-1.mdx +++ b/docs/deployments/container/gpu-speech-to-text-melia-1.mdx @@ -10,8 +10,7 @@ import melia1GpuSttDockerCompose from './assets/gpu-stt-melia-1-docker-compose.y # GPU Speech to text container (Melia 1) -Melia 1 is a multilingual, GPU-only model, generally available in Batch. For an overview of the model, see [Models](/speech-to-text/models#melia-1). Real-time (streaming) inference is available as an early preview — see [Real-time inference](#real-time-inference). - +Melia 1 is a multilingual, GPU-only model, currently available in Batch. For an overview, see [Models](/speech-to-text/models#melia-1). ## Prerequisites - [A license file or a license token](/deployments/container/licensing) diff --git a/docs/private/melia-1-realtime.mdx b/docs/private/melia-1-realtime.mdx index 2920c369..36a4d810 100644 --- a/docs/private/melia-1-realtime.mdx +++ b/docs/private/melia-1-realtime.mdx @@ -2,11 +2,13 @@ description: Melia 1 delivers code-switching for 56 languages, in Realtime. --- +import { smVariables } from '/sm-variables'; import CodeBlock from '@theme/CodeBlock'; import Tabs from '@theme/Tabs'; import TabItem from '@theme/TabItem'; import meliaRealtimePythonExample from "./assets/melia-1-rt-example.py?raw"; import meliaRealtimeJsExample from "./assets/melia-1-rt-example.js?raw"; +import melia1GpuSttDockerCompose from '../deployments/container/assets/gpu-stt-melia-1-docker-compose.yml?raw'; # Melia 1 Realtime Preview - Multilingual transcription @@ -91,7 +93,7 @@ Connect over WebSocket as described in [Realtime transcription](/deployments/con ### Multi-session containers -By default the real-time container will accept only one websocket connection at a time. To enable multiple connections, set the environment variable `SM_MAX_CONCURRENT_CONNECTIONS` to the maximum number of sessions to allow. When this is set, the `/ready` [health check endpoint](#health-service) will return true if there is a free connection available. +By default the real-time container will accept only one websocket connection at a time. To enable multiple connections, set the environment variable `SM_MAX_CONCURRENT_CONNECTIONS` to the maximum number of sessions to allow. When this is set, the `/ready` [health check endpoint](/deployments/container/cpu-speech-to-text#health-service) will return true if there is a free connection available. CPU usage scales with the number of active sessions, whereas most memory usage is shared between connections. From 752955529fbbba69ff54a5e2938ddaefc69f2f57 Mon Sep 17 00:00:00 2001 From: Georgios Hadjiharalambous Date: Fri, 11 Sep 2026 19:22:49 +0100 Subject: [PATCH 08/10] correct version docker compse --- docs/private/melia-1-realtime.mdx | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/private/melia-1-realtime.mdx b/docs/private/melia-1-realtime.mdx index 36a4d810..3e7f605a 100644 --- a/docs/private/melia-1-realtime.mdx +++ b/docs/private/melia-1-realtime.mdx @@ -110,7 +110,7 @@ This Docker Compose file will create a Speechmatics Melia 1 GPU Inference Server (assumes your `license.json` file is in the current working directory) -{melia1GpuSttDockerCompose.replace('{{ CONTAINER_VERSION }}', smVariables.latestMelia1ContainerVersion)} +{melia1GpuSttDockerCompose.replace('{{ CONTAINER_VERSION }}', 1.7.0)} ### Minimal example From e91f5f717f7a1069445ea9ed2f9fd23dae19e8be Mon Sep 17 00:00:00 2001 From: Georgios Hadjiharalambous Date: Fri, 11 Sep 2026 19:26:13 +0100 Subject: [PATCH 09/10] correct docker compose and version --- .../gpu-stt-melia-1-rt-docker-compose.yml | 36 +++++++++++++++++++ docs/private/melia-1-realtime.mdx | 8 ++--- 2 files changed, 40 insertions(+), 4 deletions(-) create mode 100644 docs/private/assets/gpu-stt-melia-1-rt-docker-compose.yml diff --git a/docs/private/assets/gpu-stt-melia-1-rt-docker-compose.yml b/docs/private/assets/gpu-stt-melia-1-rt-docker-compose.yml new file mode 100644 index 00000000..dfc38f61 --- /dev/null +++ b/docs/private/assets/gpu-stt-melia-1-rt-docker-compose.yml @@ -0,0 +1,36 @@ +version: "3.8" + +networks: + transcriber: + driver: bridge + +services: + triton: + image: speechmaticspublic.azurecr.io/sm-asr-inference-server-melia-1:{{ CONTAINER_VERSION }} + deploy: + resources: + reservations: + devices: + - driver: nvidia + ### Limit to N GPUs + # count: 1 + ### Pick specific GPUs by device ID + # device_ids: + # - 0 + # - 3 + capabilities: + - gpu + container_name: triton + networks: + - transcriber + expose: + - 8000/tcp + - 8001/tcp + - 8002/tcp + environment: + - NVIDIA_DRIVER_CAPABILITIES=all + - NVIDIA_VISIBLE_DEVICES=all + - CUDA_VISIBLE_DEVICES=0 + - SM_STREAMING=true + volumes: + - $PWD/license.json:/license.json:ro diff --git a/docs/private/melia-1-realtime.mdx b/docs/private/melia-1-realtime.mdx index 3e7f605a..656fdd77 100644 --- a/docs/private/melia-1-realtime.mdx +++ b/docs/private/melia-1-realtime.mdx @@ -8,7 +8,7 @@ import Tabs from '@theme/Tabs'; import TabItem from '@theme/TabItem'; import meliaRealtimePythonExample from "./assets/melia-1-rt-example.py?raw"; import meliaRealtimeJsExample from "./assets/melia-1-rt-example.js?raw"; -import melia1GpuSttDockerCompose from '../deployments/container/assets/gpu-stt-melia-1-docker-compose.yml?raw'; +import melia1GpuSttRtDockerCompose from './assets/gpu-stt-melia-1-rt-docker-compose.yml?raw'; # Melia 1 Realtime Preview - Multilingual transcription @@ -52,7 +52,7 @@ The Melia 1 inference server can run in two modes: _batch_ or _real-time_. The d -e CUDA_VISIBLE_DEVICES \\ -e SM_STREAMING=true \\ -p 8001:8001 \\ - speechmaticspublic.azurecr.io/sm-asr-inference-server-melia-1:${smVariables.latestMelia1ContainerVersion} + speechmaticspublic.azurecr.io/sm-asr-inference-server-melia-1:1.7.0 `} @@ -86,7 +86,7 @@ docker run --rm -it \ -e SM_ASR_MODE=rt \ -v $PWD/license.json:/license.json \ -p 9000:9000 \ - speechmaticspublic.azurecr.io/sm-asr-transcriber-melia-1:${smVariables.latestMelia1ContainerVersion} + speechmaticspublic.azurecr.io/sm-asr-transcriber-melia-1:1.7.0 ``` Connect over WebSocket as described in [Realtime transcription](/deployments/container/cpu-speech-to-text#realtime-transcription). @@ -110,7 +110,7 @@ This Docker Compose file will create a Speechmatics Melia 1 GPU Inference Server (assumes your `license.json` file is in the current working directory) -{melia1GpuSttDockerCompose.replace('{{ CONTAINER_VERSION }}', 1.7.0)} +{melia1GpuSttRtDockerCompose.replace('{{ CONTAINER_VERSION }}', '1.7.0')} ### Minimal example From 121b20b87985def8b34c80a47d5b0673c06f517c Mon Sep 17 00:00:00 2001 From: Georgios Hadjiharalambous Date: Fri, 11 Sep 2026 19:26:34 +0100 Subject: [PATCH 10/10] remove import --- docs/private/melia-1-realtime.mdx | 1 - 1 file changed, 1 deletion(-) diff --git a/docs/private/melia-1-realtime.mdx b/docs/private/melia-1-realtime.mdx index 656fdd77..6e4b5939 100644 --- a/docs/private/melia-1-realtime.mdx +++ b/docs/private/melia-1-realtime.mdx @@ -2,7 +2,6 @@ description: Melia 1 delivers code-switching for 56 languages, in Realtime. --- -import { smVariables } from '/sm-variables'; import CodeBlock from '@theme/CodeBlock'; import Tabs from '@theme/Tabs'; import TabItem from '@theme/TabItem';