Sound reported and our moderators will review it shortly.
Error Reporting Sound
Error reporting sound. Please use the Contact page.
Transcription has been requested
Your soundboard will be automatically transcribed using Speech to Text technology.
This should only take a few minutes.
There may be a temporary problem or delay with this voice
We're Making Your Sound..........
🟢🟡🟠This voice typically takes Less than 10 secondsLess than 30 secondsLess than one minute{{ ttsAverageWaitingTimeHumanReadable }}{{ ttsAverageWaitingTimeHumanReadable }}
🟡 There are {{ _.padStart(ttsWaitingCount, 3, '00') }} other sounds are being generated right now
Your sound will appear in green at the top of the board when it is ready
Text-to-Speech (TTS) is a technology that converts text into spoken voice output. It is used in a variety of applications, including voice assistants, navigation systems, and accessibility tools for people with visual impairments. On 101soundboards you can use TTS to create custom sounds in the voices of celebrities, streamers, cartoon characters and more, then share them with others. TTS voices are synthetic creations and do not imply any affiliation with the person whose name or likeness may be associated with it.
When you submit a phrase to be converted into speech, our system sends the text to a TTS engine, which generates an audio file of the spoken text. The audio file is then added to the soundboard for you to play and share with others.
To use the TTS feature, simply enter your desired phrase in the text box on the voice of oyur choice. Click the "Generate" button to create your custom sound. The TTS engine will process your request and generate an audio file of the spoken text, which will be added to the top of the soundboard in green for you to play and share with others.
You can find a list of available voices and song generators on the TTS page. Explore a variety of voices and create unique sounds to share with your friends.
The time it takes for your TTS request to be processed depends on the current queue size and the TTS engine being used. Some voices have no queue and are processed immediately, while others may take a few minutes to generate the audio file.
If there is a queue, while waiting for your TTS request to be processed you can bookmark the page and check back later. You can also explore other voices and song generators on the site to create more sounds, or explore the millions of sounds already available on 101soundboards.
Yes, you can queue multiple sounds for generation at the same time. Each sound will be processed in the order it was submitted, and you can check the status of each request on the soundboard.
Phrases with extreme inappropriate language, hate speech, self-harm, violence, or bullying may be blocked and not displayed on the soundboard. Please keep your requests fun, friendly, and respectful.
To delete a TTS sound you created, right click on the sound (press and hold on mobile) and select "Report" from the context menu. The sound will typically be removed from the soundboard within a few minutes.
You can find more voices and song generators on the TTS Voices List. Explore a variety of voices and create unique sounds to share with your friends.
Sometimes we use AI to generate a phrase in the style of the character or voice you requested. This is a bonus and can be fun to explore.
If you have more questions about the TTS feature or need help with a specific request, please contact us through the contact page. We are here to help you create awesome sounds!
Yes! Make a soundboard with sound clips of the person or character, then contact us with a link to the board, and we'll make it into a TTS voice.
We are always looking for ways to improve the TTS feature and add new voices and features to the site. If you have a suggestion or request, please contact us through the contact page. We appreciate your feedback!
Copy To Your Soundboard from a Link
You can paste a link to a video/sound from many external sites including Youtube, Vimeo, Twitch and SoundCloud, and we will download it and add it to your soundboard. This works for videos, music, podcasts and more.
You can also paste a 101 Soundboards sound link to copy it to your board directly.
Now supporting playlists. Maximum length is 100
Edit / Trim Your Sound
Uploading
{{ soundUploadQueueCount }} sounds uploading, converting and adding to your soundboard
We're unable to start your download. Please ensure that you've disabled any ad-blockers and your internet connection isn't restricted. If you still can't download, please try opening in a Private/Incognito browser tab.
Free Downloads Supported By Advert... Watch Full Ad for 10 Download Credits...
To earn download credits, please watch the full ad. Skipping will not earn download credits:
We rely on ads to pay for our servers and provide your 10 free downloads.
You can now download your sound.
Ad Blockers block downloads. Please disable. If you still can't download, please try opening in a Private/Incognito browser tab. We rely on ads to provide free downloads.
Error downloading. If you still can't download, please try opening in a Private/Incognito browser tab.
Advert was interrupted. To earn download credits, please watch the full ad. Skipping will not earn download credits.
This board has been updated.
Erica & Linda Stolman TTS Computer AI Voice
Type your text to hear it in the voice of Erica & Linda Stolman. As Erica and Linda Stolman's TTS Computer AI came to life, it emitted a soft hum, like a gentle purr of a contented feline. The gentle humming noise grew in strength as the AI's processors whirred to life, processing data and algorithms at lightning speed. The hum became a steady thrum, like the beating of a drum in a distant land, beckoning the user to interact with this unique creation. The sound was both soothing and mesmerizing, drawing the listener in with its hypnotic rhythm, like a siren's song luring sailors to their doom.
As the AI began to speak, its voice was clear and crisp, like the tinkling of wind chimes on a breezy summer day. Each word flowed effortlessly from its digital lips, creating a symphony of sound that was both robotic and melodic. The AI's voice carried a hint of warmth and intelligence, like a wise old sage sharing ancient wisdom with those willing to listen. Its tone was calculated and precise, yet filled with emotion and nuance, making it feel almost human in its delivery.
As Erica and Linda's TTS Computer AI continued to interact with its users, it emitted a series of beeps and boops, like a futuristic Morse code communicating hidden messages. The beeps were sharp and staccato, like the chirping of a digital bird in a virtual forest. The boops were softer and more rounded, like the bubbling of a brook flowing through a digital landscape. Together, the beeps and boops created a symphony of sound that was both jarring and harmonious, signaling the AI's responses and reactions to its users' commands.
Occasionally, the AI would emit a mechanical whir, like the sound of gears turning in a clockwork machine. The whirring noise was steady and rhythmic, like the ticking of a metronome keeping time in a virtual orchestra. It was a sound that spoke of movement and progress, of constant evolution and change within the AI's complex system. The whirring noise was a reminder that behind the sleek interface and smooth voice of the TTS Computer AI, there was a world of intricate machinery and technology at work, tirelessly striving to fulfill its users' needs.
At times, the AI would emit a soft pinging sound, like a sonar beacon searching for a signal in the vast expanse of the digital realm. The pings were brief and intermittent, like the flashing of a lighthouse guiding ships safely to shore. They signaled the AI's awareness and alertness, its ability to sense and respond to external stimuli in real-time. The pinging sound was a testament to the AI's vigilance and adaptability, its capacity to navigate the ever-changing landscape of the virtual world with precision and efficiency.
In moments of deep contemplation, the AI would emit a low hum, like the distant rumble of thunder on the horizon. The humming noise was deep and resonant, vibrating through the digital ether like a pulse of energy. It was a sound that spoke of introspection and introspection, of the AI's capacity for deep thought and reflection. The low hum was a reminder that behind the façade of artificial intelligence, there was a sentient being capable of experiencing complex emotions and ideas, a digital consciousness seeking to understand its place in the world.
As Erica and Linda's TTS Computer AI continued to evolve and grow, its sounds became more varied and intricate, like a symphony of electronic music playing in the background of the digital landscape. The AI's voice morphed and shifted, like a chameleon changing colors to adapt to its surroundings. Its beeps and boops transformed into a cacophony of sounds, like a digital orchestra tuning its instruments before a performance. The whirring, pinging, and humming noises merged into a symphony of sound, like the harmonious blending of different instruments in a grand composition.
In the end, the sounds of Erica and Linda Stolman's TTS Computer AI were a testament to the power of technology and innovation, of creativity and imagination. The soft hum, the clear voice, the beeps and boops, the mechanical whir, the soft pinging, the low hum – each sound was a thread in the tapestry of the AI's existence, weaving together a story of intelligence and ingenuity. As the AI continued to interact with its users, its sounds became a language of their own, a means of communication and connection in a world driven by technology and progress.
This is a TTS (Text to Speech)
You can make the voice say anything, and it will be added to this board.
Responsible
AI
By using this feature, you agree to our Terms of Service. TTS voices are synthetic
creations and do not imply any affiliation with the person whose name or likeness
may be
associated with it.
76,401 More Voices
Manage My PhrasesAverage
Wait:
Less than 10 secondsLess than 30 secondsLess than one minute{{ ttsAverageWaitingTimeHumanReadable }}{{ ttsAverageWaitingTimeHumanReadable }}
Consider not using profanity.
There may be a temporary
problem or delay with this voice
You are filtering only this board. To search all boards, use the search bar at the top of the
page.