Voice & Language Profiles
Configure a voice per language and let agents switch voice mid-call as the conversation's language changes.
If your agents need to speak more than one language, Talkr can assign a different TTS voice to each language and switch between them automatically as a call progresses — without a separate LLM or TTS provider per language.
Voice profiles use a single TTS provider for every language. All the voices you configure must come from the same provider as your organization's BYOK Voice configuration.
How it works
- You configure a Voice Profile at the organization level: one TTS provider, and a voice for each language you want to support.
- On a specific agent, you set a Primary Language and, optionally, one or more Additional Languages it should also handle.
- When a call starts, Talkr resolves the agent's primary language against your voice profile and configures the TTS voice for it.
- If additional languages are configured (and your STT provider is Deepgram), Talkr also enables Deepgram's multi-language auto-detection for the call. As the detected spoken language changes mid-conversation, the agent's TTS voice switches to match — with no added per-turn latency, since the switch context is resolved once at call setup.
If the org's voice profile doesn't have a voice for the language a workflow requests, or the profile's provider no longer matches the org's configured TTS provider, Talkr fails soft: the call proceeds on the organization's plain TTS configuration rather than failing to connect.
Configuring the Voice Profile
- Go to Model Configuration → BYOK → Voice.
- Configure your TTS provider as usual (see Voice).
- In the Voice Profile card, add an entry for each language: pick a language, then a voice available from your configured TTS provider.
- Click Save Voice Profile.
The language options offered here come directly from your TTS provider's supported language list.
Configuring an agent's languages
- Open the agent and go to Settings → Language.
- Set the Primary Language — this is the voice used from the start of the call. Only languages with a voice already configured in your organization's Voice Profile are selectable here.
- Optionally check any Additional Languages the agent should be able to switch into mid-call.
- Save the agent's settings.
Mid-call auto-switching only activates when your organization's STT provider is Deepgram — that's currently the only STT integration Talkr uses for multi-language auto-detection. With a primary language and no additional languages selected, the agent still uses the correct starting voice on every STT provider; it just won't switch mid-call.
Related
- Voice — configure your TTS provider and default voices
- Transcriber — configure your STT provider
Voice
Voice Agents use TTS (Text to Speech), which generates audio that LLMs generate during the course of a conversation. This is the audio that the end user having the conversation listens to.
Transcriber
Voice Agents use STT (Speech to Text), to transcribe what the user speaks. This transcribed speech as text goes into an LLM to generate the response that gets played out to the user.