Job Description This project requires session scheduling, recording, and delivery of complete audio files (one file per script).
The recordings will be used for Text-to-Speech (TTS) research and development purposes and will not be broadcast or used in any public or paid media.
Recording will be done on our client's recording platform.
Voice talents will be paired and responsible for scheduling recording times that accommodate their schedules.
Each script will take approximately one hour to record, and each person can record up to 10 scripts per category for a total of 20 scripts.
(See below for the categories.)
The recordings can take place over approximately one month.
There is a high probability of additional recording work in the future.
Payments will be issued for each submitted audio file after it has been reviewed and meets the requirements.
Scripts provided to you will be in the following categories: Natural Conversations Open-ended spontaneous human-human conversations touching on various topics such as current-events, movies, artists, weekend plans, friendships, challenging situations asking for support, feeling down, feeling excited, work Conversations with a specific theme, where the user's intent is to retrieve information, and where the assistant must be contextually appropriate to the theme: Sports, News, Instructional (Cooking/ Crafting/ Art) Planning tasks with the help of an assistant (trips, visits, concert tickets, restaurant bookings, shopping, creating calendar events, messaging/emailing, etc.)
Creative use cases where the user instructs the assistant to compose stories, poems and come up with jokes Customer Service Use Cases Issue with Product/Service Return a product Unauthorized purchase/access/activity Issue with Delivery Account Access Order a product or Subscribe to/Book a service Qualifications General requirements - Fluency in Spanish as it is spoken in Spain (es-ES) - Proficiency in written Spanish - Ability to deliver very expressive dialogues and professional performances, conveying the correct emotions as per the script.
The dialogues will include non verbals (e.g., laughter, sighs, gasps) to add richness and variety to the vocal performance.
- Ability to commit to submitting at least 5 recordings per week for the duration of the contract.
- Need to have a speech recording partner -- Your script-reading partner doesn't need to be in the same location (can be in any part of the world), so long as you/him/her speak the same language/ accent.
Nice-to-haves (Preference will be given to talent with these qualifications.)
- Experience working with scripts (for example, voice-over work, audiobook reading, acting) - Experience with professional speech recording Equipment requirements - PC or Laptop with the latest operating system on either Microsoft Windows, Mac OS or Linux (Redhat or Debian-based distros) and the most recent release version of Google Chrome installed; and - Webcam and microphone with high-quality audio - High-speed, password protected internet connection.
Performing work over an open wireless network connection is NOT allowed Technical Requirements for submitted recordings - File Type: Unprocessed, RAW .wav - Noise Floor: -60dB or better - Quality: Naturally minimized background noise, breaths, and clicks (no plugins).
Recording must be done in a quiet environment with minimal background noise.
Application Process: - Interested talent should submit a proposal and their bid for recording each approximately one-hour script.
As a guideline, the average rate for one recorded script is $32.
You may propose a rate based on that guideline and your experience and skills.
Please keep in mind that no Upwork fees will be deducted from your earnings, meaning you will receive the full bid amount if it is accepted.
If you know someone who would be interested in participating in the two person dialogue with you, please let us know in your proposal.
- A script attachment has been added to this posting as well.
As an optional step, you may also submit a recording of yourself reading the script.