# Voice & Speech (/docs/agents/voice-agents/voice-and-speech)
> For the complete documentation index, see [llms.txt](https://docs.tandemcx.ai/llms.txt). Any page on this site can be fetched as Markdown by appending `.md` to its URL.
The **Voice & Speech** card, on the editor's **Conversation** step, controls how the agent
sounds and how it shares the conversation. A voice is required; the rest have defaults that
work.
The card itself holds **Allow barge-in**, **Words before interrupting**, the voice cascade and
**Caller language recognition**. The slider button in the card's header opens a **Speech
Tuning** panel with the rest: **Turn detection**, a **Thinking cue**, **Voice activity
detection**, and the selected provider's own voice character settings.
The whole card is **Voice only**. It is absent, and so is the requirement for a voice, on an
agent with **Voice** switched off in its
[Channels](/docs/agents/voice-agents/persona#channels).
## Allow barge-in
Barge-in decides what happens when the caller starts talking while the agent is talking.
It's **on** by default, which means caller speech interrupts the agent mid-sentence — the
agent stops, and listens.
Leave it on for nearly everything. Callers interrupt naturally, and an agent that talks over
them feels broken.
Turn it off only when the agent must be heard to the end: a required disclosure, a recording
notice, a legal script. With barge-in off, the caller can't stop the agent mid-utterance, so
keep those utterances short.
### Words before interrupting
Below the switch is **Words before interrupting** — how much the caller has to say before their
speech actually stops the agent. Leave it empty and any sound interrupts, which is what every
agent does today: a cough, a throat-clear, or an "mhm" of agreement cuts the agent off
mid-sentence. Set it to `3` and none of those three stops the agent, because none of them is
three words.
**It's a trade, not a free filter.** With a threshold set, an interruption is detected from the
caller's first transcribed *words* rather than from the moment speech is heard. That slows
barge-in on genuine interruptions too — the caller who really did want to cut in waits for those
words to be transcribed as well. Setting `3` to stop cough-interruptions is accepting a slower
stop for everyone.
Two things that aren't obvious:
- **`1` is not the same as empty.** Empty uses speech detection; `1` still switches to transcribed
words, so it adds the delay without filtering anything. `2` is the smallest setting that actually
filters a one-word backchannel.
- **It only applies while the agent is talking.** An agent waiting for the caller still takes a
single word as the start of a turn, so a one-word answer to a question is never swallowed.
It refines barge-in rather than replacing it, so it does nothing at all while **Allow barge-in** is
off — an agent playing a disclaimer to the end has no interruption for a word count to refine. The
field stays visible and editable there anyway, so the value is still yours to change. 1–10.
## Choosing a voice
The voice is chosen through four dependent dropdowns, each one populated by the choice above
it. Pick them in order:
### Voice Provider
Which text-to-speech service produces the audio.
### Language
The language and regional variant the voice speaks, such as `en-US`. This is also the
language the agent's speech is synthesized in, so it should match what you expect callers to
hear — and, unless you override it, the language callers are transcribed in. A mismatch
here doesn't just sound wrong; it mishears the caller. See
[Recognition language](#recognition-language) below.
### Voice
The specific named voice. Providers offer several per language.
### Voice Engine
The synthesis quality tier — typically **Standard**, **Neural**, **Generative**, or
**Long-form**, depending on what the chosen voice supports. Higher tiers sound more natural;
they can also add a little latency.
For an ElevenLabs voice, the engine is the ElevenLabs model:
- **Flash v2.5 (fastest)** — the lowest latency.
- **Multilingual v2 (highest quality)** — offered for English voices only, because this model
does not accept a language code.
- **v3 Conversational (expressive)** — ElevenLabs' Eleven v3 model for real-time speech. It
has more latency than Flash (ElevenLabs quotes about 280 ms against about 75 ms).
- **v4 Turbo (most expressive)** — ElevenLabs' Eleven v4 model for real-time speech (ElevenLabs
quotes about 100 ms). Offered for English voices only.
v3 Conversational and v4 Turbo read only **Stability** from the voice character settings.
Changing any level clears the levels below it, so switching language means re-picking the
voice and engine, because a voice from one language isn't valid in another.
Listen to a candidate voice on a real phone call, not on your laptop speakers. Phone audio is
narrowband, and voices that sound distinct in a browser can sound nearly identical over a call.
Some models produce speech themselves rather than handing text to a separate voice service.
When you select one of those on the **Brain** card, the provider is fixed by the model and
the dropdown is replaced with a note saying so.
## Multiple languages
Multiple languages is enabled per organization. If the **Voice & Speech** card has no
**Languages** section, it isn't on for yours yet. If it is turned off after an agent was given
more languages, the rows stay as saved and calls use the primary language only. You can remove
a row there (its greeting parts go with it), but you can't add or change one.
One agent can hold a call in up to five languages: a primary voice and up to four more, each with
its own voice. Calls start in the primary language. When the caller speaks another of the
languages, or asks for it, the agent switches its voice and replies in that language.
### Add a language
In the **Languages** section, the voice you chose above becomes the **Primary language**. Select
**Add Language** to add a row, then fill it in like the primary voice: provider, language, voice,
and engine. Rows can use different providers. A row is titled with its language, voice, and
provider, and carries an **Incomplete** badge until all four are picked. The delete button on a
row removes it.
Rules the editor enforces:
- **Five languages in all**, the primary plus four. At the limit **Add Language** is disabled,
with the reason beside it.
- **One voice per language family.** `en-US` and `en-GB` can't both be on one agent, because the
agent knows both as English and switches by language name. The picker hides a dialect of a
language already listed. If a draft got there another way, **Save** fails with
` is the same language as . Use one voice per language family.`
- **A language needs a speech recognition peer.** A language that Amazon Transcribe can't
transcribe while streaming is disabled in the picker with "no speech recognition". If the
primary voice's language is one of these, **Add Language** is disabled. See
[Recognition language](#recognition-language) for which languages these are.
- **Not available with a model that speaks in its own voice**, because that model takes no extra
voices.
Removing a language that a greeting part is written in is refused until you change or remove that
part. See [Greeting](/docs/agents/voice-agents/persona#greeting).
### How the agent switches
The caller is transcribed in the language the call is in. Amazon Transcribe listens for that
one language, so speech in another of the agent's languages reaches the agent as the
nearest-sounding words in the current one: "Können Sie bitte Deutsch sprechen?" can arrive as
"Kenzie Bitterdeutschprechen." The agent is told to read the caller's words for what they sound
like, and switches the call in one of two ways:
- It replies in another of its languages. The reply is spoken in that language's voice, and the
call continues in it (see [Which voice speaks](#which-voice-speaks)). The language a reply
starts in is the one that counts: a reply that repeats a line in a second language speaks that
line in the second language's voice and keeps the call where it was.
- It calls the built-in `switch_language` tool when the caller asks for another language or
names one in answer to a menu. See [Switch
language](/docs/agents/voice-agents/built-in-tools#switch-language).
From then on the caller is transcribed in the new language. The agent is told not to switch on
a single name or filler word in another language.
A caller who starts speaking another language without asking can be misheard on that first
turn, because it is transcribed in the current language. A menu in the greeting makes the
switch dependable, for example "For Spanish, say español."
The call's language follows the call. It carries across a transfer to another agent, a return from
a workflow, and a reconnect after a dropped call.
### Which voice speaks
The agent's voice always follows the language of what the agent says. Each reply is spoken in
the voice for the language the reply is written in. If the agent replies in English, the
English voice reads it, even when the caller was last heard in German. If the reply is in a
language the agent has no voice for, the voice stays as it was. A reply in another configured
language also becomes the call's language, so the next turn continues in it unless the caller
switches again.
### The greeting and the thinking cue
A multi-language agent can play its greeting in several voices, one part per language. See
[Greeting](/docs/agents/voice-agents/persona#greeting). A returning-caller greeting you wrote
yourself is one block of text, so it's spoken in a single voice: the one for the language the call
is in when it starts. The [thinking cue](#thinking-cue) works the same way: the text you write
is spoken in the voice of the language the call is currently in, and isn't translated.
### If a voice fails
Each language keeps its own failover. An ElevenLabs voice that fails falls back to an Amazon Polly
voice for the same language.
If the provider's key can't be used when a call starts, a language that has a Polly voice (Spanish,
for example) is spoken by that Polly voice from the first word.
Some languages have no Polly voice: Croatian, Greek, Filipino, Hindi, Indonesian, Malay, Slovak,
Ukrainian, and Vietnamese. A language from that list, as an additional language on a voice from
another provider, is left out of the call when the provider's key can't be used as the call
starts. The call continues in the languages that remain. The primary voice is never
left out: if its language is one of these, an English voice speaks it. If the key is accepted
at the start and fails during the call, an English voice reads that language aloud. The
language's row in the **Languages** section shows a note when this applies.
### Custom vocabulary
Tandem builds a [vocabulary](/docs/agents/voice-agents/vocabulary) for each language's speech
recognition when you publish. Calls use it only once every language's vocabulary is ready. Until
then calls run without recognition tuning, in every language. The prompt part of vocabulary
isn't affected.
## Recognition language
Next to the voice cascade is **Caller language recognition** — a separate setting from the
voice's own language. Leave it blank and it follows the voice's language automatically, so a
Spanish-language voice transcribes a Spanish-speaking caller correctly with no extra setup.
Override it for the asymmetric case: a caller who speaks one language to an agent that
answers in another — for example, a caller speaking Spanish to an agent whose voice, and
therefore replies, are in English. Without an override, that call would be transcribed as
English and the agent would consistently mishear the caller.
The dropdown offers every Transcribe **streaming** language, which is a narrower set than the
languages Polly can speak — the list is the whole supported set, so anything absent from it has
no streaming counterpart. See [Troubleshooting](#troubleshooting) for what happens when a
voice's language is one of those.
With more than one language, the dropdown is replaced by a list of the agent's languages, and
you can't override it. The caller is transcribed in whichever of them the call is in. Adding
the first extra language clears an override you had set, and a notice on the card says what it
was. Removing the last extra language brings the dropdown back, set to follow the voice.
## Voice character
Once a voice is chosen, the **Advanced** disclosure adds provider-specific tuning for how
that voice sounds. Every field is optional — leave it blank and the vendor's own default
applies, so a config that predates these fields keeps behaving exactly as it did before.
### Polly voice character
Shown when the selected voice provider is Polly.
- **Speaking rate** — a prosody keyword (`x-slow`, `slow`, `medium`, `fast`, `x-fast`) or a
percentage from `20%` to `200%` inclusive. Anything else is rejected when you save. For
example, to slow an agent down, set it to `90%`.
- **Volume** — a prosody keyword (`silent`, `x-soft`, `soft`, `medium`, `loud`, `x-loud`) or a
decibel offset such as `+2dB` or `-6dB`. The offset's format is checked when you save, but
its range isn't — an out-of-range value is only caught by Polly itself, when it fails to
synthesize speech, not when you save the agent.
### ElevenLabs voice character
Shown when the selected voice provider is ElevenLabs.
- **Speed** — `0.7` to `1.2`.
- **Stability** — `0` to `1`.
- **Similarity boost** — `0` to `1`.
- **Style exaggeration** — `0` to `1`.
- **Speaker boost** — on or off.
- **Text normalization** — `auto`, `on`, or `off`. This is ElevenLabs' own handling of digit
runs, currency, and dates in the text it speaks (for example, expanding `$5` to "five
dollars"), not a setting Tandem applies itself.
On the **v3 Conversational** and **v4 Turbo** engines only **Stability** applies. The other
settings are saved but ElevenLabs ignores them for those models.
### Pronunciation dictionaries
**Pronunciation dictionaries** sits under the ElevenLabs voice character settings. It lists the
pronunciation dictionaries in your organization's ElevenLabs workspace, read with the key on
your ElevenLabs integration, and applies the ones you pick to every line the agent speaks. Use
one for a product name, a brand, a drug name, or a surname the voice gets wrong. For example, a
dictionary with the rule `GLaDOS` → `ˈɡlædoʊz` makes the agent say the name that way every
time.
- You can pick up to 3. Once 3 are picked, the rest of the list is disabled until you remove
one.
- Each call uses a dictionary's latest version. An edit you make to a dictionary in ElevenLabs
reaches calls within five minutes, with no change to the agent.
- You create and edit dictionaries in ElevenLabs. Tandem lists them and applies them; it
doesn't edit them.
- If your ElevenLabs integration isn't connected, the field says so and lists nothing. Connect
it, then use the refresh button beside the field.
A dictionary is not checked when you save. Each call checks it with ElevenLabs instead. If
ElevenLabs reports it doesn't exist (it was deleted, or it lives in a different ElevenLabs
workspace from the connected key), that call goes ahead without it. If ElevenLabs doesn't answer
in time, the call can't get the latest version, so it goes ahead without that dictionary too.
Once the list of your dictionaries has loaded, one still saved on the agent but no longer in your
workspace shows as `(not found in ElevenLabs)`, so you can remove it. While the list is loading,
fails to load, or ElevenLabs isn't connected, a saved dictionary shows as its id alone.
Dictionaries don't apply everywhere:
- **v3 Conversational** and **v4 Turbo** don't support pronunciation dictionaries. The field
says so, and the saved dictionaries are ignored on those engines.
- If ElevenLabs fails during a call and the agent switches to its Amazon Polly standby voice,
Polly doesn't apply the dictionaries.
Through the API, the field is `slots.tts.settings.elevenlabs.pronunciationDictionaries` in the
agent configuration, and
[List pronunciation dictionaries](/docs/api/voice-agents/providers/listPronunciationDictionaries)
returns the ids. An entry is `{ "id": "…" }`, or `{ "id": "…", "versionId": "…" }` to pin one
version instead of following the latest. A pinned dictionary is still checked at call time and
skipped if ElevenLabs reports it doesn't exist. If ElevenLabs doesn't answer in time, a pinned
dictionary is used anyway, since the call already has its version. The pinned version itself
can't be checked (ElevenLabs has no way to list a dictionary's versions), so pin only a version
you know exists.
Each call on **v3 Conversational** or **v4 Turbo** holds one ElevenLabs dialogue session for as
long as the call lasts, from a pool separate from your ElevenLabs plan's usual concurrency
limit. When every session in that pool is in use, a new call can't open one and falls back to an
Amazon Polly voice. Check your plan's dialogue session limit in ElevenLabs before moving a busy
agent to this engine.
## Thinking cue
When the agent calls a tool — looking up an account, running a workflow — there's a pause
while that work happens. By default the line is silent, and callers take a silent line for a
dropped call.
The **Thinking cue** is a short phrase spoken while a tool call runs:
```
One moment while I check that…
```
Keep it generic. It plays before _any_ tool call, so it can't reference a specific lookup.
Keep it short, too. It should fill a gap, not become the pause. On a new agent, leaving the
field empty means no cue at all, which is fine for agents with no tools.
**Typing a phrase replaces whatever the agent had**, including an audio-file cue configured
through the API — this box only ever writes a spoken cue. Clearing the box and saving now
removes the saved cue entirely (spoken or audio-file); the agent goes back to a silent pause
during tool calls.
## Turn detection
Turn detection is how the agent decides the caller has **finished** — the other half of barge-in,
which decides when they've started. There are two strategies, chosen under **Strategy**.
**Silence timers** is what every agent has always used, and it stays the default. The agent waits
for the caller to stop making sound, then waits a little longer in case they resume. It's cheap and
completely predictable, and it can't tell a thinking pause from a finished sentence, because all it
measures is silence. Everything under [Voice activity
detection](#voice-activity-detection) below tunes it.
**Smart turn model** runs a small speech model that predicts whether the caller is done from how
they sounded — falling intonation, a phrase that resolved — rather than from silence alone. A caller
who says "so my account number is… hold on, let me find it" gets waited for instead of answered
into.
### When smart is worth it
Reach for it when callers are being cut off mid-sentence and raising **End-of-turn grace** has made
every other reply feel sluggish. Silence timers cannot separate those two cases: one number sets
both the tolerance for a pause and the delay on every turn. A model separates them.
Stay on silence timers when turn-taking is already fine, when calls are short and transactional
("say or enter your account number"), or when you need turn-taking to behave identically on every
call — a model makes a prediction, and predictions vary.
**Smart mode is not free.** Each concurrent call using it costs roughly **19 MB of memory and
one processing thread** on the voice service, plus about 20 ms of model loading while the call
is being set up. That's why it's a per-agent choice: turn it on for the agents whose
conversations need it, not across the board.
### What stops applying in smart mode
**End-of-turn grace is not used in smart mode.** The model decides end-of-turn itself and has no
equivalent setting, so a grace value you tuned is still stored and still shown, and never
applied. **Silence limit is not its replacement.** That's the ceiling on a turn the model got
*wrong* (3,000 ms by default, against the 800 ms silence timers wait by default), not the wait
on a normal one. If you switch strategy, expect the turn-taking numbers you tuned to stop
meaning what they meant.
The rest of the VAD fields still do their jobs: **Silence before reply**, **Detection threshold**,
**Speech start delay**, and **Minimum volume** are about detecting _sound_, not about deciding a
turn is over. What changes is that the silence timer now only marks where the caller stopped
speaking — the model, not the clock, decides whether that was the end.
### Smart-mode tuning
Choosing the smart strategy reveals three more fields, each optional — leave one blank and the
model's own default applies. Once any of them holds a value it stays on screen on silence timers
too, so you can see and clear what you set; it just isn't read there. Only smart mode reads them.
**Silence limit** — the longest silence the model will sit through before calling the turn over
regardless of what it predicted. A backstop on a mispredicted turn, not the normal wait.
100–10,000 ms; default 3,000.
**Pre-speech padding** — how much audio from just _before_ the caller started speaking the model
gets to hear. The start of a phrase is part of what it reads. 0–5,000 ms; default 500.
**Audio window** — the longest stretch of caller audio the model scores at once; anything older is
trimmed off the front. 1–30, default 8. **Note the unit:** this field is in **seconds** where the
other two are milliseconds, and fractions are allowed.
Change one at a time and test on a real call, the same as the VAD fields.
## Voice activity detection
Voice activity detection (VAD) is how the agent hears that the caller is talking, and — on the
silence-timer strategy — how it decides they've finished a turn. Five fields tune it, each
individually optional: leave any of them blank and the underlying voice pipeline's own default
applies.
**Silence before reply** — how long the caller must stop talking before the agent counts the turn
as over. Milliseconds; default 200.
**End-of-turn grace** — after **Silence before reply** elapses, a further window in which the
caller may resume talking without the agent replying. Milliseconds; default 600.
**The two add together.** What a caller experiences as "how long until the agent replies" is
`Silence before reply + End-of-turn grace`, not either one alone. With both at their defaults,
that's 0.2s + 0.6s = **0.8s total**. Raising or lowering only **Silence before reply** and
expecting turn-taking to feel different is why this used to feel untunable — the grace window kept
adding its own 0.6s regardless of what **Silence before reply** was set to.
**End-of-turn grace applies to the silence-timer strategy only.** With [Turn
detection](#turn-detection) set to the smart turn model, the model decides end-of-turn and this
field is never read — the card says so under the field, and **Silence limit** is not a replacement
for it. See [What stops applying in smart mode](#what-stops-applying-in-smart-mode).
**Detection threshold** — how confident the detector must be that the sound it hears is speech.
Raise it on noisy lines so background noise doesn't register as talking; lower it for soft
speakers who get treated as silence. 0–1; default 0.7.
**Speech start delay** — how long incoming speech must continue before it counts as the caller
starting a turn, rather than a stray sound. Milliseconds, 0–2,000; default 200.
**Minimum volume** — the quietest sound level still treated as speech rather than background
noise. 0–1; default 0.6.
Leave all five empty unless you have a specific problem to solve. When you do change them,
change one at a time and test on a real call — several of them interact, and moving more than
one at once makes it impossible to tell which one helped. Clearing a field and saving now
restores the pipeline's own default for it; it no longer leaves the old value in place.
**Raising End-of-turn grace to tolerate a caller's mid-sentence pause delays every reply by the
same amount.** No VAD setting distinguishes a thinking pause from a finished sentence. It only
measures silence. A grace window generous enough to survive "um, let me check…" makes every
other reply that much slower too. Raise it only as far as the callers you're tuning for actually
need. The smart turn model exists to remove this trade-off, so if you are fighting it, read
[Turn detection](#turn-detection) above.
## Limits and defaults
| Setting | Value |
| ------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------------------------------------------------- |
| **Allow barge-in** | On by default |
| **Words before interrupting** | Optional; 1–10; any sound interrupts when absent; ignored while barge-in is off |
| Voice Provider | Required, unless the model produces audio itself or **Voice** is off in Channels |
| Language | Required once a provider is chosen |
| Voice | Required once a language is chosen |
| Voice Engine | Required once a voice is chosen |
| **Thinking cue** | Optional; empty means silence on a new agent |
| Turn detection **Strategy** | Silence timers by default |
| Turn detection **Silence limit** | Optional; smart only; 100–10,000 ms; default 3,000 |
| Turn detection **Pre-speech padding** | Optional; smart only; 0–5,000 ms; default 500 |
| Turn detection **Audio window** | Optional; smart only; 1–30 **seconds**, fractions allowed; default 8 |
| VAD **Silence before reply** | Optional; positive whole number of ms, no upper bound; default 200 |
| VAD **Detection threshold** | Optional; 0–1; default 0.7 |
| VAD **Speech start delay** | Optional; 0–2,000 ms; default 200 |
| VAD **Minimum volume** | Optional; 0–1; default 0.6 |
| VAD **End-of-turn grace** | Optional; 0–10,000 ms; default 600; **not read on the smart turn model** |
| **Languages** | Optional; up to 5 in all (the primary plus 4 added); one voice per language family; each needs a Transcribe streaming language |
| **Caller language recognition** | Optional; one of the listed Transcribe streaming languages; follows the voice's language when absent; cleared when a second language is added |
| Polly **Speaking rate** | Optional; a prosody keyword or `20%`–`200%`; Polly's own default applies when absent |
| Polly **Volume** | Optional; a prosody keyword or a decibel offset (range unchecked at save); Polly's own default applies when absent |
| ElevenLabs **Speed** | Optional; 0.7–1.2; the vendor's own default applies when absent |
| ElevenLabs **Stability** | Optional; 0–1; the vendor's own default applies when absent |
| ElevenLabs **Similarity boost** | Optional; 0–1; the vendor's own default applies when absent |
| ElevenLabs **Style exaggeration** | Optional; 0–1; the vendor's own default applies when absent |
| ElevenLabs **Speaker boost** | Optional; on or off; the vendor's own default applies when absent |
| ElevenLabs **Text normalization** | Optional; `auto`, `on`, or `off`; the vendor's own default applies when absent |
| ElevenLabs **Pronunciation dictionaries** | Optional; up to 3, each once; ignored on v3 Conversational and v4 Turbo |
| Clearing a saved cue, VAD, recognition-language, or voice-character value | Restores the default; takes effect on save |
## Troubleshooting
**The agent talks over the caller.** Confirm **Allow barge-in** is on. If it is, check **Words
before interrupting** — a threshold set there means short utterances no longer stop the agent at
all. If that's empty too, the caller's speech isn't registering: lower the VAD **Detection
threshold** so quieter speech counts.
**The agent interrupts itself, or stops for background noise.** The detector is treating
noise as speech. Raise the **Detection threshold**.
**The agent still stops for a cough or an "mhm".** Set **Words before interrupting** to 2 or 3.
Accept that genuine interruptions get a little slower in exchange — see the warning under that
field.
**Interrupting the agent now takes a beat too long.** Lower **Words before interrupting**, or
clear it. Any value at all, including `1`, switches interruption detection over to transcribed
words; only an empty field uses plain speech detection.
**The agent cuts callers off mid-sentence.** Raise the VAD **Silence before reply** so it waits
longer before deciding the turn is over.
**The agent cuts in when the caller pauses mid-sentence to think.** Raise **End-of-turn grace** —
it's the window that lets the caller resume without the agent jumping in. Remember it delays every
reply by the same amount, so raise it only as far as your callers need. If you've reached the point
where every reply feels slow, that's what the smart turn model under **Turn detection** is for.
**I switched to the smart turn model and my tuning stopped mattering.** **End-of-turn grace** isn't
read in that mode — the model decides end-of-turn instead. **Silence limit** is a backstop on a
mispredicted turn, not the same knob. Switch back to silence timers and your saved values apply
again, unchanged.
**There's an awkward gap before every reply.** Lower **Silence before reply** or **End-of-turn
grace** — together they set the total wait. If the gap only happens on some replies, it's a tool
call rather than turn-taking — set a **Thinking cue**.
**The line goes silent for several seconds mid-call.** That's a tool call with no cue. Set a
**Thinking cue**.
**A custom audio thinking cue turned into a spoken phrase.** Someone typed text into **Thinking
cue** and saved, which overwrites an audio-file cue set through the API. Re-apply the audio cue
through the API, and leave the box empty on future saves from this card.
**I cleared the Thinking cue and it's gone, but I wanted to keep the audio-file cue.** Clearing
and saving now removes whatever cue was set, spoken or audio-file — see the warning above.
Re-apply the audio cue through the API if you cleared it by mistake.
**The voice sounds robotic.** Try a higher **Voice Engine** tier for the same voice, if the
voice offers one.
**Save fails asking for a voice.** The cascade is incomplete. All four levels must be filled.
Re-pick from the top, since changing one level clears the ones below it.
**The agent speaks correctly but the transcript reads like a different language, or the
agent keeps misunderstanding a caller who's speaking the voice's own language.** The voice's
language may have no Transcribe streaming counterpart, so the caller is being transcribed as
English regardless of what the voice speaks. This is a known gap for **Bulgarian, Hungarian,
Tamil, Turkish, Welsh, and Icelandic** — the six voice languages with no streaming peer.
Arabic is supported in both variants: Modern Standard Arabic and Gulf Arabic each transcribe
in their own variant. The **Caller language recognition** dropdown is
the definitive list: a language that isn't in it is a language callers can't be transcribed
in. There's no indicator on the voice cascade itself today, so if the agent's voice is one of
these languages, set **Caller language recognition** explicitly to the closest supported
language you expect callers to use, or choose a different voice language.
**The agent didn't switch when the caller started speaking another language.** The caller is
transcribed in the call's current language, so a short phrase in another language can read as
unrelated words, and the agent is told not to switch on a single word. Offer a menu in the
greeting ("For Spanish, say español") so the agent can switch on the caller's answer with
`switch_language`.
**A language is missing from the language picker.** Either it has no Transcribe streaming
language, or another voice on the agent already covers its language family (`es-ES` and `es-US` are
both Spanish).
**Spanish callers are recognized as the wrong regional Spanish.** Recognition follows the
voice. An ElevenLabs Spanish voice (`es`) is recognized as Spanish from Spain (`es-ES`), and an
Amazon Polly `es-US` voice as Spanish from the United States (`es-US`). Pick the voice whose
region matches your callers.
**Calls run without my vocabulary in one language.** Calls use custom vocabulary only once
every language's vocabulary is ready, so one language still building, or failed, holds back all of
them. The status line on the [Vocabulary](/docs/agents/voice-agents/vocabulary) card shows the
first language that isn't ready.
## Next steps