For a workflow that has audio,

1. Within DeepHow, navigate to the Editor and select a workflow that has the new banner.
2. Select Edit Video and select confirm to continue.
3. We encourage users to make any Cuts to the workflow first before utilizing the text to speech tool to ensure you have all the video contents correct
4. Navigate to the Text-to-Speech button at the top after you have made your cuts.
5. First, an overview of this page.
   • On the left hand side, it will take each sentence of the video and generate a text block for each.
   • On the right hand side, you will see the video to reference.
   • On the bottom, you will see a Voice selection option and a Generate TTS button, along with the original video on a timeline with the TTS audio blocks underneath.
6. To start, select the Voice selection button and pick a voice.
   • You have the option to preview the voice before selecting.
   • When you find the right voice, select it.
   • This voice will be used to generate all your audio in the video.
   • Please note, that if you select a voice after the audio has been generated, you will have to generate the audio again as you can not have two different voices in a video.
7. Now its time to adjust the transcription via the text blocks.
   • We recommend that each line has it’s own text block. This is because you will have more control of where the line falls on the video
   • You have the option to delete a block of text by selecting the delete button.
   • You can also customize the text in each block
   • You can change the speed of the voice-generated speech by selecting 1x speed on the text block, and then selecting either .5x, .75x, 1.25x, 1.5x,
   • You can also drag and drop text blocks to customize the order of the transcription
   • You can highlight between a text block and add a new text block
   • You can leave text blocks empty if desired. These are used as space holders where you don’t want audio in the video.
   • When you select Generate on an audio block, the AI Stephanie will take a second depending on how long the text is, and then generate the audio. To do this for the full video, you can select generate
8. You can align the audio and video by dragging the text blocks on the timeline
9. Once ready, you can select the play button to hear the text-to-speech.
10. You can play the whole video within the editor to listen to how everything came out.
11. When you select Save, you will get this message that Stephanie is creating the voice to your text. You can go back to the editor to continue editing other videos or wait for Stephanie to finish.
12. Check your work before you Select Save. If you leave before saving, you will get a warning notification that the changes you made will be lost.
13. Once done, you will see the new reflected transcription and be able to continue on to segmentation.

     

For a workflow without audio

1. Within DeepHow, navigate to the Editor and select a workflow that has the new banner.
2. You will see no transcription and you will see this option to select either creating a voice over or text to speech. Select Text to speech.
3. We would advise that you first do any cuts to the video before you add text to speech. Once ready with your video, navigate to Text to speech.
4. You will see no text blocks on the left hand side as there is no transcription yet. Our first step is to select our voice that we would like to generate our text. You can preview the voice with the audio icon here.
5. You will also see that there is no audio track listed underneath the video on the timeline so as you add text blocks they will start appearing in blue.
6. Now we'll start adding text blocks. You can type the text here or copy and paste the text that you want to add to the script.
7. We wouldn't recommend that you add a big block of text to the text to speech.
8. This is because it'll be harder to navigate and align the text to your video so we would recommend creating a block for each sentence.
9. Once ready, select generate all TTS to see where the text aligns on the video.
10. You can minimize the timeline on the bottom and drag the text to increase the length of the block.
11. Sometimes your audio track may go past your video track. We would suggest that your audio track ends at the same time as the video. If you select save when there is more audio track than video, you will get an error.
12. When you are finished adjusting, select save.
13. Now you will see the voice has been generated. You can make more edits or select X to continue to segmentation.

• For more information on how to upload videos with no audio, read our article here 

• Leave your feedback on our community post 

FAQ

How do I use the DeepHow Text-to-Speech feature?

A: Upload a workflow in DeepHow’s Editor, select edit the video, and select Text-to-Speech. Learn more in our step-by-step guide and watch our video above. Read our article here on how to use TTS for existing workflows.

Why would I use the Text-to-Speech feature?

A: Maybe… 

• You have a loud environment to record audio 
• You want a standard voice across all your content 
• You don’t like the sound of your voice or are nervous 
• You want to reduce background noise when discussing a process 
• It takes a lot of tries to get the voice correct when recording a process

Which languages are supported by the Text-to-Speech feature?

A: The languages supported are English, Chinese, French, German, Finnish and Spanish. Preview and choose from different voice types to add a touch of personality to your content.

Is the TTS feature available on all apps?

A: Text-to-Speech is available on DeepHow’s Editor on the browser. Once published, users will be able to hear the audio-generated TTS.

Can I have different voices for different text blocks?

A: No, you will have to select one voice for the full transcription. 

Can you preview the voice before generating all text blocks? 

A: Yes, you can preview the voice by selecting the language button and selecting the audio icon.

Can I adjust the speed of the audio?

A: Yes, you can adjust the pace of the voice to match your preferred tempo.

Can my English text be translated within the TTS editor? 

A: No, the tool does not translate your text at this time. If you select a Spanish voice for your English text, it will be spoken with a Spanish accent. 

When will other languages be available in TTS?

A: We will add languages to the TTS feature in future iterations. If you have a language you would like to see, please submit your feature requests in our Community.

Does TTS apply to the attachment I add to my workflows?

A: No, TTS only applies to the transcription within the video.

Can I customize the voices in TTS or am I limited the options provided?

A: Currently we have two to four voices in each language to select from.  If you have a voice preference you would like to see, please submit your feature requests in our Community.

If I have audio on my workflow before using TTS, will I lose the audio once I save TTS?

A: Yes, once you save on the TTS tool, you will replace your audio track with the TTS track.

Can I use voiceover after I use the TTS tool? 

A: You will lose your TTS content when you utilize the voiceover tool. 

Leave your feedback on our community post