Bonjour! Hallo!​ Now Supporting French and German

At Rime, we're all about making AI conversations more representative of real people across the world. So we're excited to announce that Mist v2, our flagship conversational text-to-speech (TTS) model, now supports French and German voices.​

Bridging Language Gaps with Mist v2

Mist v2 has been at the forefront of delivering ultra-fast, lifelike speech for real-time applications. With the addition of French and German, we're enabling developers and enterprises to create more authentic, multilingual voice experiences.​

Whether you're building customer support agents, interactive voice response (IVR) systems, or any voice-driven application, Mist v2's new language capabilities ensure your solutions resonate with a broader audience.​

Built on a Foundation of Linguistic Expertise

Our commitment to linguistic diversity and authenticity is evident in our data collection and model training processes. We brought in a large group of French and German native speakers to our downtown San Francisco studio to record many hours of full duplex speech data. This approach ensures that our models, including Mist v2, capture the nuances of real human speech across different languages and cultures.​

Get Started Today

Explore the new French and German voices on our dashboard and via our API now. See how Mist v2 can elevate your voice applications.!

👉 Try French and German in Mist v2 Now

Code Samples

French

Here's a sample API data param:

{"text": "bonjour!", "speaker": "juliette", "modelId": "mistv2", "lang": "fra"}

Valid speaker params are:

alois
juliette
marguerite

Here's a quick CURL request example for MP3 over JSON:

curl --request POST \
 --url https://users.rime.ai/v1/rime-tts \
 --header 'Accept: application/json' \
 --header 'Authorization: Bearer <authorization>' \
 --header 'Content-Type: application/json' \
 --data '{
 "speaker": "juliette",
 "text": "Bonjour, comment vas-tu aujourd'hui?",
 "modelId": "mistv2",
 "lang": "fra",
 "samplingRate": 22050,
 "speedAlpha": 1.0,
 "audioFormat": "mp3",
 "reduceLatency": false,
 "phonemizeBetweenBrackets": false,
 "pauseBetweenBrackets": false,
}'

German

Here's a sample API data param:

{"text": "hallo!", "speaker": "klaus", "modelId": "mistv2", "lang": "ger"}

Valid speaker params are:

klaus
frieda
karolina
maximilian
amalia

Here's a quick CURL request example for MP3 over JSON:

curl --request POST \
 --url https://users.rime.ai/v1/rime-tts \
 --header 'Accept: application/json' \
 --header 'Authorization: Bearer <authorization>' \
 --header 'Content-Type: application/json' \
 --data '{
 "speaker": "klaus",
 "text": "Hallo, wie geht es Ihnen heute?",
 "modelId": "mistv2",
 "lang": "ger",
 "samplingRate": 22050,
 "speedAlpha": 1.0,
 "audioFormat": "mp3",
 "reduceLatency": false,
 "phonemizeBetweenBrackets": false,
 "pauseBetweenBrackets": false,
}'

We can't wait to hear what you build!