ElevenLabs TTS Operator

The ElevenLabs TTS LOP generates high-quality, natural-sounding speech from text using the ElevenLabs API. It utilizes WebSockets for low-latency audio streaming directly into a TouchDesigner CHOP, making it suitable for real-time applications. It also supports saving the generated audio to local files and recalling previously generated audio from ElevenLabs’ history or local cache.

API Key Required

This operator requires a valid API key from ElevenLabs. Ensure you have an account and have entered the key in the parameters.

Requirements

• Python Packages:
  • websockets: For streaming audio.
  • requests: For fetching voices and history.
  • numpy: For audio data handling.
  • soundfile: For saving audio files (install via pip install soundfile).
  • Install these via the ChatTD Python Manager.
• ChatTD Operator: Required for dependency management, API key storage (optional), and asynchronous task execution. Ensure the ChatTD Operator parameter on the ‘About’ page points to your configured ChatTD instance.
• ElevenLabs API Key: A valid API key from your ElevenLabs account.

Input/Output

Inputs

• Input DAT (input_message, optional): If Text Input Source is set to input_dat, the text content of this DAT is used for synthesis.

Outputs

• Output CHOP (null_audio_out -> store_output): Contains the streamed or loaded audio data (mono, 24kHz).
• History DAT (history_dat): Internal table logging local generation jobs initiated by this operator.
• API History DAT (api_history_dat): Internal table mirroring recent history items fetched from the ElevenLabs API.
• Local History DAT (local_history_dat): Internal table listing locally saved audio files generated by this operator.
• Voices DAT (voices_dat): Internal table listing available voices fetched from the API.
• Saved Audio Files: Audio can be saved locally (WAV or OGG) in the specified Save Folder.
• Saved Metadata Files: JSON files containing details about each saved generation are stored in a metadata subfolder within the ElevenLabs config directory.

Parameters

Page: ElevenLabs

ElevenLabs API Key (Apikey) op('tts_elevenlabs').par.Apikey Str

Default:

None

Get API Key (Getapikey) op('tts_elevenlabs').par.Getapikey Pulse

Default:

None

Fetch Available Voices (Getvoices) op('tts_elevenlabs').par.Getvoices Pulse

Default:

None

Voice (Voiceid) op('tts_elevenlabs').par.Voiceid Menu

Default:

None

Model (Modelid) op('tts_elevenlabs').par.Modelid Menu

Default:

eleven_multilingual_v2

Options:

eleven_multilingual_v2, eleven_turbo_v2, eleven_monolingual_v1, eleven_multilingual_v1, eleven_flash_v2_5

Stability (Stability) op('tts_elevenlabs').par.Stability Float

Default:

0

Range:

0 to 1

Slider Range:

0 to 1

Similarity Boost (Similarityboost) op('tts_elevenlabs').par.Similarityboost Float

Default:

0

Range:

0 to 1

Slider Range:

0 to 1

Use Speaker Boost (Usespeakerboost) op('tts_elevenlabs').par.Usespeakerboost Toggle

Default:

Off

Text Input Source (Textinput) op('tts_elevenlabs').par.Textinput Menu

Default:

parameter

Options:

parameter, input_dat

Text to Speak (Text) op('tts_elevenlabs').par.Text Str

Default:

None

Start Streaming (Stream) op('tts_elevenlabs').par.Stream Pulse

Default:

None

Stop Streaming (Stop) op('tts_elevenlabs').par.Stop Pulse

Default:

None

Clear Stored Audio (Clearaudio) op('tts_elevenlabs').par.Clearaudio Pulse

Default:

None

Status (Status) op('tts_elevenlabs').par.Status Str

Default:

None

Active (Active) op('tts_elevenlabs').par.Active Toggle

Default:

None

Playing (Playing) op('tts_elevenlabs').par.Playing Toggle

Default:

None

Start Playback (Playstart) op('tts_elevenlabs').par.Playstart Pulse

Default:

None

Enable Playback Control (Play) op('tts_elevenlabs').par.Play Toggle

Default:

Off

Play on Result (Playonresult) op('tts_elevenlabs').par.Playonresult Toggle

Default:

Off

Check Credits (Checkcredits) op('tts_elevenlabs').par.Checkcredits Pulse

Default:

None

Credit Info (Creditinfo) op('tts_elevenlabs').par.Creditinfo Str

Default:

None

Page: Playback

Audio Device Settings Header

Active (Audioactive) op('tts_elevenlabs').par.Audioactive Toggle

Default:

On

Driver (Driver) op('tts_elevenlabs').par.Driver Menu

Default:

default

Options:

default, asio

Device (Device) op('tts_elevenlabs').par.Device Menu

Default:

default

Options:

default, {0.0.0.00000000}.{d7b929aa-ec27-4f96-bd39-78d6a8c2044a}||Out_1-2_(MOTU_M_Series)||1, {0.0.0.00000000}.{0f011619-4954-432e-b132-2653e7136839}||LEN_P32u-10_(NVIDIA_High_Definition_Audio)||2, {0.0.0.00000000}.{10c42639-6ae9-4121-ad66-04d4f850c7e6}||VBMatrix_In_7_(VB-Audio_Matrix_VAIO)||3, {0.0.0.00000000}.{372fb62b-07aa-4580-b062-2c6adba187e7}||Out_3-4_(MOTU_M_Series)||4, {0.0.0.00000000}.{3f0d389a-29d3-415d-a9ba-20773d212885}||VBMatrix_In_6_(VB-Audio_Matrix_VAIO)||5, {0.0.0.00000000}.{55a6de81-1fa3-4b2e-bced-71880481eb62}||VBMatrix_In_8_(VB-Audio_Matrix_VAIO)||6, {0.0.0.00000000}.{6f55253a-13a1-40f9-9f80-762ce203338b}||VBMatrix_In_3_(VB-Audio_Matrix_VAIO)||7, {0.0.0.00000000}.{762727ea-ca22-466b-9f2a-c6ca2d6d9648}||VBMatrix_In_1_(VB-Audio_Matrix_VAIO)||8, {0.0.0.00000000}.{7c163c65-5cda-4fbd-971b-27e1c3b471f0}||F13NA_(NVIDIA_High_Definition_Audio)||9, {0.0.0.00000000}.{ad8837d6-c905-4a72-adcb-7018ee7baab3}||CABLE_Input_(VB-Audio_Virtual_Cable)||10, {0.0.0.00000000}.{b3dcef2a-cec9-4686-ae78-79573bf3df3e}||VBMatrix_In_4_(VB-Audio_Matrix_VAIO)||11, {0.0.0.00000000}.{b6526f47-8c31-48d5-8b30-489196c56a6b}||Headphones_(iLoud_Micro-Monitor)||12, {0.0.0.00000000}.{cc818f2b-580f-4915-ad18-53412c5378fa}||VBMatrix_In_5_(VB-Audio_Matrix_VAIO)||13, {0.0.0.00000000}.{deaebd99-08fa-47eb-bac0-864267bd9522}||DELL_S2415H_(NVIDIA_High_Definition_Audio)||14, {0.0.0.00000000}.{df2e555b-b4ef-4656-921f-6272a16b48a7}||VBMatrix_In_2_(VB-Audio_Matrix_VAIO)||15

Volume (Volume) op('tts_elevenlabs').par.Volume Float

Default:

1

Range:

0 to 1

Slider Range:

0 to 1

Save Audio To File Header

Auto Save To Disk (Autosavetodisk) op('tts_elevenlabs').par.Autosavetodisk Toggle

Default:

Off

Output Folder (Folder) op('tts_elevenlabs').par.Folder Folder

Default:

"" (Empty String)

Base Name (Name) op('tts_elevenlabs').par.Name Str

Default:

None

File Type (Filetype) op('tts_elevenlabs').par.Filetype Menu

Default:

wav

Options:

wav, mp3, ogg

Auto Version (Autoversion) op('tts_elevenlabs').par.Autoversion Toggle

Default:

On

Save Current Audio (Savefile) op('tts_elevenlabs').par.Savefile Pulse

Default:

None

CHOP Outputs Header

Output 1 (CHOP) (Out2chop) op('tts_elevenlabs').par.Out2chop Menu

Default:

playback

Options:

playback, full_audio

Page: History

Update History (Fetchhistory) op('tts_elevenlabs').par.Fetchhistory Pulse

Default:

None

Audio to Load (Generationid) op('tts_elevenlabs').par.Generationid Menu

Default:

None

Select From (Selectfrom) op('tts_elevenlabs').par.Selectfrom Menu

Default:

all

Options:

all, all_local, project_only

Display (Display) op('tts_elevenlabs').par.Display Menu

Default:

job_id

Options:

job_id, text, timestamp, local_file, name

Load Audio File (Loadfile) op('tts_elevenlabs').par.Loadfile Pulse

Default:

None

Enable History Playback (Playbackfromhistory) op('tts_elevenlabs').par.Playbackfromhistory Toggle

Default:

Off

Playing History Item (Playinghistory) op('tts_elevenlabs').par.Playinghistory Toggle

Default:

None

Loaded Audio File Path (Audiofilepath) op('tts_elevenlabs').par.Audiofilepath Str

Default:

None

Page: Callbacks

Callbacks Header

Callback DAT (Callbackdat) op('tts_elevenlabs').par.Callbackdat DAT

Default:

ChatTD_callbacks

Create Callbacks (Createcallbacks) op('tts_elevenlabs').par.Createcallbacks Pulse

Default:

None

onGenerate (Ongenerate) op('tts_elevenlabs').par.Ongenerate Toggle

Default:

Off

onDone (Ondone) op('tts_elevenlabs').par.Ondone Toggle

Default:

Off

onPlayStart (Onplaystart) op('tts_elevenlabs').par.Onplaystart Toggle

Default:

Off

onPlayEnd (Onplayend) op('tts_elevenlabs').par.Onplayend Toggle

Default:

Off

onSave (Onsave) op('tts_elevenlabs').par.Onsave Toggle

Default:

Off

Textport Debug Callbacks (Debugcallbacks) op('tts_elevenlabs').par.Debugcallbacks Menu

Default:

Full Details

Options:

None, Errors Only, Basic Info, Full Details

Available Callbacks:

• onGenerate
• onDone
• onPlayStart
• onPlayEnd
• onSave

Example Callback Structure:

def onGenerate(info):
# Called before streaming starts
# info: {'job_id', 'text', 'voice_id', 'model_id'}
print(f"Starting generation job: {info.get('job_id')}")

def onDone(info):
# Called when streaming finishes or errors
# info: {'job_id', 'status', 'message', 'duration', 'text', 'voice_id', 'model_id'}
print(f"Generation finished: {info.get('status')} ({info.get('duration')}s)")

def onPlayStart(info):
# Called when playback starts
# info: {'job_id', 'audio_duration', 'text'}
print(f"Playback started for job: {info.get('job_id')}")

def onPlayEnd(info):
# Called when playback ends
# info: {'job_id', 'audio_duration', 'text'}
print(f"Playback finished for job: {info.get('job_id')}")

def onSave(info):
# Called after audio is saved successfully
# info: {'job_id', 'filepath', 'filename', 'folder', 'duration', 'text'}
print(f"Audio saved to: {info.get('filepath')}")

Page: About

Bypass (Bypass) op('tts_elevenlabs').par.Bypass Toggle

Default:

Off

Show Built-in Parameters (Showbuiltin) op('tts_elevenlabs').par.Showbuiltin Toggle

Default:

Off

Version (Version) op('tts_elevenlabs').par.Version Str

Default:

None

Last Updated (Lastupdated) op('tts_elevenlabs').par.Lastupdated Str

Default:

None

Creator (Creator) op('tts_elevenlabs').par.Creator Str

Default:

None

Website (Website) op('tts_elevenlabs').par.Website Str

Default:

None

ChatTD Operator (Chattd) op('tts_elevenlabs').par.Chattd OP

Default:

None

Clear Log (Clearlog) op('tts_elevenlabs').par.Clearlog Pulse

Default:

None

Convert To Text (Converttotext) op('tts_elevenlabs').par.Converttotext Toggle

Default:

Off

Usage Examples

Basic Streaming and Playback

1. Enter your ElevenLabs API Key.
2. Pulse ‘Fetch Available Voices’ and select a voice from the ‘Voice’ menu.
3. Select a desired ‘Model’.
4. Keep ‘Text Input Source’ as ‘parameter’.
5. Enter text in ‘Text to Speak’.
6. Ensure ‘Play on Result’ is ON.
7. Pulse ‘Start Streaming’. Audio should stream in and play automatically.

Streaming from Input DAT

1. Set ‘Text Input Source’ to ‘input_dat’.
2. Create a Text DAT and connect it to the first input of the ElevenLabs operator.
3. Enter text into the Text DAT.
4. Pulse ‘Start Streaming’.

Saving Audio Automatically

1. Ensure ‘Auto Save To Disk’ is ON.
2. Specify an ‘Output Folder’.
3. Optionally change the ‘Base Name’ (use $JOBID for uniqueness) and ‘File Type’.
4. Run streaming as usual (e.g., Pulse ‘Start Streaming’).
5. After successful streaming, the audio and metadata files will appear in the specified folder.

Recalling and Playing History

1. Pulse ‘Update History’ to fetch recent items from the API and local storage.
2. Select an item from the ‘Audio to Load’ menu.
3. Pulse ‘Load Audio File’. Check ‘Status’ and ‘Loaded Audio File Path’.
4. Enable ‘Enable History Playback’.
5. Use the standard ‘Start Playback’ pulse to play the loaded audio.

Technical Notes

• Streaming: Uses WebSockets for low-latency audio delivery (PCM format, 24kHz sample rate).
• API Key: Stored securely in a local configuration file (elevenlabs_config.json) within the ChatTD environment or project folder.
• Dependencies: Requires websockets, requests, numpy, soundfile.
• Asynchronous: All API communication (streaming, voice fetch, history fetch, credit check) happens asynchronously via ChatTD to avoid blocking TouchDesigner.
• Saving: Audio saved locally requires the soundfile library. Metadata is saved in JSON format.
• History: Fetches history from the ElevenLabs API (limited items) and scans local save folder for generated files.
• Playback: Uses an internal Audio Device Out CHOP configured via the Playback page parameters.

Related Operators

• ChatTD: Provides core services like dependency management and asynchronous task execution.