fAIr Voices: Consent, Control, Compensation
Are you a company that handles voice actor audio files? Please join our pledge.
fair voices PLEDGE
#fAIrVoices: Training, Consent, Control, Compensation, Security, Sourcing
NAVA supports the right of all performers to protect themselves from unauthorized usage through digital cloning of their voices, likenesses, and performances. NAVA understands however, that the power of protection is often in the control of the producers or clients that they work for. NAVA and it’s members desire to empower themselves by obtaining assurances from those they work for that the people and institutions they trust with their recorded performances will not break that trust by engaging in any unauthorized usage.
The companies listed below have provided those assurances.
#fAIrVoices: Training, Consent, Control, Compensation, Security, Sourcing
TRAINING:
Any company which requires the signing of terms of service, will have clearly outlined terms of service or company policies, which prohibit the unauthorized use of a person’s data and sound files for the creation and/or training of Synthetic voice models created using AI and/or Machine Learning.
CONSENT:
All users of our company’s services have the right to actively consent to the use of their voice, image, name and/or likeness for the creation, training, and modeling of Synthetic voice models, digital clones, and replicas created using AI and/or Machine Learning. This includes, but is not limited to, the use of sound files, video, or other media containing a user’s biometric data.
CONTROL:
All users of our company’s services have reasonable control of the work created using their voice, image, name, and likeness.
If a synthetic voice model or digital double is created, users have reasonable control over the types of work that can be generated by their synthetic voice model, digital clone, or replica. This control may be project-specific, and the parameters of this control should be decided upon in advance.
COMPENSATION:
All users should be fairly compensated, according to a mutually agreed upon rate and rate structure if a company ever wishes to use their voice or sound files for the creation, training, and modeling of Synthetic voice models created using AI and/or Machine Learning.
Note: Some examples of industry standard rates are from SAG-AFTRA, GVAA, or other industry wide accepted rate guides applicable to the location or country you are working in.
SECURITY:
Our company agrees to use good faith efforts to prevent any files of recordings or performances stored in digital format containing a user’s voice or likeness from unauthorized access by third parties, and if such files are stored in “the cloud” Client agrees to utilize services that offer safeguards through encryption or other “up-to date” technological means from unauthorized third party access.
SOURCING:
Our company agrees that all files used for the training and creation of Synthetic voice models created using AI and/or Machine Learning will be fairly sourced, with the active consent and knowledge of the person the files are sourced from.
Ashley Bigbie
Owner
www.audiobeeproductions.com
AUDIOBEE TOS
We have strong anti-AI values on our company. In our website’s TOS, section 4.1, we outline that AI scraping and training is prohibited on our site. Any page that has narrator vocal samples on public display is protected with a CloudFlare “Are you human?” Verification. In our service agreements and contracts when hiring narrators for an audiobook, we have multiple clauses where we, the Producer, and our audio engineers are prohibited from using the audio files for any AI or machine learning purposes. Our service agreements and contracts with our clients (authors) also prohibits the client from using our audio files or our narrators voices for any AI or machine learning.
SEE TRAINING
SEE TRAINING
SEE TRAINING
SEE TRAINING
SEE TRAINING
Sonja Smith
Business Dev., VO Casting
www.backstage.com
Armin Hierstetter
Chief Executive Officer
We do not train AI on our platform: https://www.bodalgo.com/en/tac
Please see: Training
We provide, where applicable, a link to the GVAA rate guide during posting a casting.
We provide security on our website for all files uploaded to Bodalgo.
We do not source AI voices for any purpose.
Liz Atherton
Founder and CEO
Cissy Jones
CEO/Founder
www.ethovox.ai
Per the Ethovox talent agreement: Each Party may terminate this Agreement at any time, with or without cause, upon thirty (30) days’ notice. If such termination is by Ethovox without cause, Ethovox shall upon such termination pay Talent all unpaid, undisputed amounts due for the Services completed prior to notice of such termination. If such termination is by Talent without cause, Talent agrees to the continued and ongoing usage of the Deliverables by Ethovox for all licenses and sub-licenses existing at the time of termination as between Ethovox and its customers in which Talent previously agreed to the use of the Deliverables.
Per the Ethovox talent agreement: Each Party also may terminate this Agreement at any time, with or without cause, upon thirty (30) days’ notice. If such termination is by Ethovox without cause, Ethovox shall upon such termination pay Talent all unpaid, undisputed amounts due for the Services completed prior to notice of such termination. If such termination is by Talent without cause, Talent agrees to the continued and ongoing usage of the Deliverables by Ethovox for all licenses and sub-licenses existing at the time of termination as between Ethovox and its customers in which Talent previously agreed to the use of the Deliverables.
Per our talent agreement: Permitted Use only as specified in SAG-AFTRA CATEGORY II Corporate/Educational and Non- Broadcast track usage on a customer-by-customer, project-by-project basis, to be approved by Talent.
Compensation per SAG-AFTRA Agreement Line #3 and Line #7: Recording Session Fees-
The following rates are in effect as of the date of this Agreement and shall increase indexed to the Corporate/Educational & Non-Broadcast Contract, Category II rates: ● 1st hour: $563.00 ● Each additional ½ hour: $148.00 In addition, Developer agrees to pay the then-current rate in the Co-Ed (currently, eighteen and seventy-five hundredths percent (18.75%)) on all gross compensation as contributions to the Plans (“Plan Contribution”). Plan Contributions are due at the time the Performer(s)’s compensation is paid and shall be paid directly to the Plans. Performers shall be paid not less than the minimum applicable fees due hereunder not later than thirty (30) days after services have been rendered.
Security language is incorporated into each talent contract.
Ethovox’s agreements with clients and talent mandate that any use of its AI-generated voice models be subject to the informed consent of talent, limited to the permitted use agreed between client and talent, subject to the ongoing control by talent over usage, and that talent must be compensated according to SAG-AFTRA guidelines.
LA Casting
Casting Director
LA Casting does not train, create, or use AI or synthetic voices.
LA Casting does not train, create, or use AI or synthetic voices.
LA Casting does not train, create, or use AI or synthetic voices.
LA Casting does not train, create, or use AI or synthetic voices.
LA Casting uses good faith efforts to prevent any files of recordings or performances stored in digital format containing Talent’s voice or likeness from unauthorized access by third parties, and if such files are stored in “the cloud” we utilize services that offer safeguards through encryption or other “up-to date” technological means from unauthorized third party access.
LA Casting does not train, create, or use AI or synthetic voices.
Chris Ciulla
CEO
www.leonardoaudio.com
Leonardo Audio does not train, create, or use AI or synthetic voices.
Leonardo Audio does not train, create, or use AI or synthetic voices.
Leonardo Audio does not train, create, or use AI or synthetic voices.
Whenever rate quotes are needed by clients, SAG-AFTRA and GVAA rates are provided.
Leonardo Audio will use good faith efforts to prevent any files of recordings or performances stored in digital format containing Talent’s voice or likeness from unauthorized access by third parties, and if such files are stored in “the cloud” we utilize services that offer safeguards through encryption or other “up-to date” technological means from unauthorized third party access.
Leonardo Audio does not train, create, or utilize AI or digital replica voices.
Reece Bridger
Company Director
Adam Fritz
CEO
The “Permitted Uses” are Pozotron’s business use, including use of Data for: (i) providing the Services hereunder, (ii) research, development, trouble-shooting/de bugging and testing of the Services, and (iii) improvement of the functionality of the Services. For clarity, improvement of the functionality of the Services includes making anonymized electronic copies of sentences from the Transcript (“Sentences”) and copies of the corresponding segments in the recordings in which those Sentences are recorded (“Audio Segments”), for the purpose of electronically comparing the Sentences and the Audio Segments in order to improve the accuracy of the Service’s ability to compare voices to text.
Company shall not use Customer’s Data for the purposes of mapping/replicating the performance elements of human narration and creating an artificial likeness of a voice provided by Customer as part of the Services hereunder, without the prior consent of the Customer and the person whose voice is used in the Recording (the “Consenting Narrator”).
Pozotron does not train, create, or utilize AI or digital replica voices.
Pozotron does not train, create, or utilize Ai or digital replica voices.
Pozotron uses industry best practices to secure our database, network, and other elements of our IT infrastructure. Users are strongly encouraged to enable 2 Factor Authentication on their accounts to maximize the security of user data.
Pozotron does not train, create, or utilize Ai or digital replica voices. Any datasets used for training of other Pozotron features including AI or Machine Learning are used in accordance with Pozotron’s Terms of Service and/or the applicable dataset licenses for any dataset acquired by Pozotron.
Tim Friedlander
CEO
www.soundbox.la
soundBOX does not train, create, or use AI or synthetic voices.
soundBOX does not train, create, or use AI or synthetic voices.
soundBOX does not train, create, or use AI or synthetic voices.
Whenever rate quotes are needed by clients, SAG-AFTRA and GVAA rates are provided.
All soundBOX studios use good faith efforts to prevent any files of recordings or performances stored in digital format containing Talent’s voice or likeness from unauthorized access by third parties, and if such files are stored in “the cloud” we utilize services that offer safeguards through encryption or other “up-to date” technological means from unauthorized third party access.
soundBOX Group does not train, create, or utilize AI or digital replica voices.
Faith Falkner
Co-Founder and Narrative Producer
Sublight Games agrees not to enter into any other agreements or contracts on behalf of the Contractor which utilize all or any part of any of the recordings, performances, arts, assets, or other Works of the Contractor without the Contractor’s prior knowledge and explicit written consent.
Sublight Games does not submit data for synthetic voice models, digital clones, replicas, or AI or machine learning. Our contractors are welcome to consent to their likeness being utilized in this way elsewhere if so desired.
Sublight Games standard contracts all include the latest NAVA AI rider.
Sublight Games specifically agrees not to sell or transfer ownership of all or part of any of the recordings, performance, arts, assets, or other Works of the Contractor to any third party without the Contractor’s prior knowledge and explicit written consent.
Sublight Games negotiates fair non-union and SAG-E rates which may be revisited at any time at the request of the Contractor.
Sublight Games takes all precautions to keep data secure via end-to-end and at-rest encryption. Production accounts must use two-factor authentication.
Sublight Games does not train, create, or utilize AI or digital replica voices.
Crystal Storm
Co-Founder
The Tales Network often holds casting calls, and produces audio dramas, fiction podcasts, and TTRPG shows. We do not condone the usage of AI/Machine Learning. We will never sell Voice Actor audio files to a third party, nor will use synthetic or doubled voices.
Absolutely no consent is given for any third parties to use any Tales of the Forgotten Fiction Network productions, including but not limited to artwork and sound production, to train their AI/Machine Learning Models.
TOS
See: Training
TOS
See: Training
TOS
See: Training
TOS
See: Training
TOS
See: Training
Rolf Veldman
CEO
VOICE123 TOS
We have internal processes that are not covered in the ToS. Example:
– Every new job poster that runs a TTS project will have to follow the questionnaire that NAVA created. This is not in the ToS, but is a manual process we manage.
Mathias Lynnerup Jakobson
CEO & Partner
Jay O’Connor
CEO
www.voices.com
Refer to our Job Agreements, which provide protective language that prohibits any AI-voice related uses of Talent audio files without written consent from Talent.
Also refer to the Voices TOS, which prohibits Voices to use Talent Content (such as auditions and demonstrations) for any uses related to AI voice without express written consent from the Talent.
Voices.com TOS
Refer to our Job Agreements, which provide protective language that prohibits any AI-voice related uses of Talent audio files without written consent from Talent.
Also refer to the Voices TOS, which prohibits Voices to use Talent Content (such as auditions and demonstrations) for any uses related to AI voice without express written consent from the Talent.
Voices.com TOS
Refer to Voices Job Agreements, which clearly specify both permitted and non-permitted uses of Talent Content involved in the given job.
In addition, for a description of Voices commitment to Ethical AI, see the following web page:
For all projects at Voices, job agreements or statements of work explicitly spell out compensation for the project as well as both permitted and non-permitted uses of Talent Content.
In addition, for a description of Voices commitment to Ethical AI, see the following web page:
For a description of Voices commitment to Ethical AI, see the following web page:
Also, refer to our responses and links above under “Consent,” “Control,” and “Compensation.”
Kelly Moscinski
Owner & Head of Casting
www.voicecaster.com
The Voicecaster does not train, create, or use AI or synthetic voices.
The Voicecaster does not train, create, or use AI or synthetic voices.
The Voicecaster does not submit data for synthetic voice models, digital clones, replicas, or AI or machine learning. Our contractors are welcome to consent to their likeness being utilized in this way elsewhere if so desired. The Voicecaster strives to work with all of our clients to sign the latest NAVA AI rider.
Whenever rate quotes are needed, The Voicecaster uses industry standard rates from SAG-AFTRA, GVAA, or other industry wide accepted rate guides applicable to the location or country you are working in.
The Voicecaster will use good faith efforts to prevent any files of recordings or performances stored in digital format containing Talent’s voice or likeness from unauthorized access by third parties, and if such files are stored in “the cloud” we utilize services that offer safeguards through encryption or other “up-to date” technological means from unauthorized third party access.
The Voicecaster does not train, create, or use AI or synthetic voices.
Recordings may not be used to develop an artificial voice model, now or in the future, on any platform. Possessing voiceover audio via auditions or audio provided as part of a paid for project does not entitle any users of the platform to reproduce, mimic, or otherwise replicate the voice of the individual who created it. Users are expressly forbidden from doing so, either directly or indirectly, without prior written permission from the voice artist. Such consent must specify the type of AI Technology intended for use, such as AI Voice Cloning or AI Voice Model Training. No AI Technology, including AI Voice Cloning or AI Voice Model Training, may be employed unless all parties have provided written consent outlining the project’s scope and the specific AI Technology to be used. Any users of the platform agree to indemnify VoicesUS against all claims, damages, liabilities, settlements, legal fees, and expenses arising from any violation of this provision.
See: Training
Voices.UK TAC
Voices.US TOS
See: Training
Voices.UK TAC
Voices.US TOS
US Rate Info (GVAA Rate Guide)
UK Rate Info (Gravy For the Brain Rate Guide)
Kevin West
Chief Executive Officer
www.VOPlanet.com
VO Planet TAC
Regarding Artificial Intelligence, Digital Voice Cloning, Use of Digital Voices
VOPlanet, LLC does not work with, have agreements with or condone the use of voiceover created by artificial intelligence (AI) in any manner. VOPlanet, LLC will delete jobs posted by AI companies soliciting use of voice work for digital cloning purposes. VOPlanet, LLC does not allow the use of artificial intelligence voiceover creators to create accounts, buy memberships or audition for VOPlanet, LLC projects. VOPlanet, LLC does not accept offers for the use of our voice files for use in creation of digital voice clones or AI voices. VOPlanet, LLC voiceover talent are freelance contractors and are free to negotiate fair compensation for use of their voice for digital cloning as they see fit if those transactions are conducted off our Site, directly between the voice buyer and the voiceover talent.
Please see: Training
Please see: Training
Please see: Training
Please see: Training
Please see: Training
Miles Chicoine
Chief Executive Officer
www.voquent.com
