magpie-tts-multilingual Model by NVIDIA

Nemotron Speech supports gRPC and HTTP APIs.

Inference using HTTP API

List available voices:

Bash

curl -s https://877104f7-e885-42b9-8de8-f6e4c6303969.invocation.api.nvcf.nvidia.com/v1/audio/list_voices \
     -H "Authorization: Bearer $NVIDIA_API_KEY"

Synthesize speech. Replace text, language, and voice values as needed.

Bash

curl -s -X POST https://877104f7-e885-42b9-8de8-f6e4c6303969.invocation.api.nvcf.nvidia.com/v1/audio/synthesize \
     -H "Authorization: Bearer $NVIDIA_API_KEY" \
     -F "text=this audio is generated from nvidias text to speech model" \
     -F "language=en-US" \
     -F "voice=Magpie-Multilingual.EN-US.Aria" \
     -F "encoding=LINEAR_PCM" \
     -F "sample_rate_hz=44100" \
     --output audio.wav

Inference using gRPC API

Instructions below demonstrate usage of magpie-tts-multilingual model using Python gRPC client. For full API details refer gRPC API Reference.

Prerequisites

You will need a system with Git and Python 3+ installed.

Install Nemotron Speech Python Client

Bash

pip install -U nvidia-riva-client

Download Python Client

Download Python client code by cloning Python Client Repository.

Bash

git clone https://github.com/nvidia-riva/python-clients.git

Run Python Client

Open a command terminal and execute below command to synthesize audio from the example text. If you have generated the API key, it will be auto-populated in the command.

Bash

python python-clients/scripts/tts/talk.py \
    --server grpc.nvcf.nvidia.com:443 --use-ssl \
    --metadata function-id "877104f7-e885-42b9-8de8-f6e4c6303969"  \
    --metadata authorization "Bearer $NVIDIA_API_KEY" \
    --language-code en-US \
    --text "this audio is generated from nvidia's text to speech model" \
    --voice "Magpie-Multilingual.EN-US.Aria" \
    --output audio.wav

List of available voices can be obtained using below command.

Bash

python python-clients/scripts/tts/talk.py \
    --server grpc.nvcf.nvidia.com:443 --use-ssl \
    --metadata function-id "877104f7-e885-42b9-8de8-f6e4c6303969"  \
    --metadata authorization "Bearer $NVIDIA_API_KEY" \
    --list-voices

Support for gRPC clients in other languages

Nemotron Speech uses gRPC APIs. Proto files can be downloaded from Nemotron Speech gRPC Proto files and compiled to target language using Protoc compiler. Example Nemotron Speech clients in C++ and Python languages are provided below.