Integration

Typecast integration

Typecast (by Neosapience) is an AI text-to-speech API. Turn text into expressive speech with its 600+ built-in voices or your own cloned voices (emotion presets, context-aware smart emotion, pitch, tempo and loudness), get word and character timestamps for subtitles, compose one audio file from several voices and pauses, find voices by filters or by a plain-language description, clone a voice from a recording (only your own voice or a speaker's explicit, verifiable consent, per Typecast's terms: a voice is biometric personal data), and check the plan's credits. Speech is billed at 1 credit per character of text; audio made on the free plan may not be used commercially and must credit Typecast. Not exposed: the streaming speech route (chunked audio for live playback; text_to_speech returns the same audio as a file) and the deprecated v2 voice and v1 clone and delete routes. Authentication is an API key from the Typecast API console, sent in the X-API-KEY header.

Connect Typecast How integrations work

Ask it in chat

Say what should happen.
The workflow builds itself.

When this happens, do that in Typecast.Built from the 12 Typecast tools below, on the connection you add once.

What you can do with Typecast

Each of these is one operation on one endpoint, available both as a workflow step and as a tool an AI agent can call on its own. Nothing bundles several operations into one action, so what a run did stays readable afterwards.

Each badge shows the default agent permission of an endpoint (allowed: an agent calls it without asking; ask first: you are asked before an agent calls it, when the agent or chat asks permission for sensitive actions; never: no agent may call it), which every user can change endpoint by endpoint in their credential settings.

Most endpoints run on the free plan. A few, mainly the ones that publish on your behalf, need a paid plan.

Custom voice, 5 operations

create_instant_cloneCreate instant clonePOSTAgents: ask firstClones a voice at once from one WAV or MP3 recording (5 to 150 seconds, 25 MiB at most) and returns its uc_ voice_id. A voice is biometric personal data: only your own voice or a speaker's explicit, verifiable consent. Uses a custom voice slot.
create_professional_cloneCreate professional clonePOSTAgents: ask firstStarts training a professional voice from one WAV or MP3 recording (5 minutes to 3 hours, 1 GiB at most, 16 kHz or more); returns voice_id, status training. Biometric personal data: your own voice or a speaker's explicit, verifiable consent only.
delete_custom_voiceDelete custom voiceDELETEAgents: ask firstDeletes one of your custom voices and frees its slot; the voice can no longer be used for speech. If a professional clone is still training, its training is canceled.
get_custom_voiceGet custom voiceGETAgents: ask firstReturns one custom voice you own with its status (pending, training, completed or failed) and the error when failed. Use it to follow a professional clone started with create_professional_clone (training can take up to two hours).
list_custom_voicesList custom voicesGETAgents: ask firstLists the custom voices your account owns (instant and professional clones) with their status (pending, training, completed or failed), the error when failed, model, source and creation time.

Speech, 3 operations

compose_speechCompose speechPOSTAgents: ask firstBuilds one audio file from up to 50 ordered segments: speech segments (each with its own voice, model and settings) and silent pauses. Bills 1 credit per character of speech text (2000 at most in total); pauses are free.
text_to_speechText to speechPOSTAgents: ask firstGenerates speech from up to 2000 characters of text with one voice and returns the audio file (WAV by default, MP3 through output.audio_format). Bills 1 credit per character. Find a voice_id with list_voices or recommend_voices.
text_to_speech_with_timestampsText to speech with timestampsPOSTAgents: ask firstGenerates speech like text_to_speech but answers JSON: the audio as base64 text in audio, plus word and character start and end times in seconds for subtitles. Bills 1 credit per character. granularity limits the timestamps.

Voice, 3 operations

get_voiceGet voiceGETAgents: ask firstReturns one built-in (tc_) or completed custom (uc_) voice: localized names, gender, age, use cases, the models it supports with their emotion presets, and a preview audio URL when one exists.
list_voicesList voicesGETAgents: ask firstLists the built-in voices and your completed custom voices (custom first): voice_id, localized names, gender, age, use cases, the models each supports with their emotions, and a preview URL. Filter by model, gender, age, use case or type.
recommend_voicesRecommend voicesGETAgents: ask firstRecommends up to 10 voices for a plain-language description of the style, mood, language or use case wanted, best match first, with voice_id, name and score. Use get_voice for a candidate's models and emotions.

Subscription, 1 operation

get_subscriptionGet subscriptionGETAgents: ask firstReturns your Typecast API plan (free, lite, plus or custom), its monthly credits and the credits used, the concurrency limit, custom and professional voice slot limits, and professional clone training runs used.

Use Typecast from Claude Code, Cursor or Codex

Connect Typecast in LiveContext, create an API key in Settings, then add the LiveContext MCP server to your client. Its Typecast operations become tools the client can call, next to every other integration you connected.

Claude Code

claude mcp add --transport http livecontext https://livecontext.ai/mcp --header "X-API-Key: YOUR_API_KEY"

Codex (~/.codex/config.toml)

[mcp_servers.livecontext]
url = "https://livecontext.ai/mcp"

[mcp_servers.livecontext.http_headers]
"X-API-Key" = "YOUR_API_KEY"

Cursor, Claude Desktop and other JSON configs

{
  "mcpServers": {
    "livecontext": {
      "type": "http",
      "url": "https://livecontext.ai/mcp",
      "headers": {
        "X-API-Key": "YOUR_API_KEY"
      }
    }
  }
}

Set up the MCP server step by step

How to connect Typecast

Create your workspace
Sign up free. Nothing to install.
Connect Typecast once
Add your Typecast connection in Settings. Every tool from this integration reuses it, and secrets are encrypted at rest and visible only to you.
Describe the job in chat
The workflow builds itself in front of you and picks the endpoints it needs. Drag the steps yourself on the canvas whenever you would rather be explicit.