Generate sound effects using your voice as a guide

Last updated on Jun 26, 2025

Learn how to use the Voice to sound effects feature to generate sound effects using your voice as a guide.

Try it in the app
Generate sound effects by using text prompts in a few simple steps.

With Voice to sound effects (beta), you can use your voice to guide the generation of sound effects. When you combine your recorded voice, which indicates when the sound effect should occur in the timeline and how loud it should sound, along with a text prompt, Firefly refers to the energy and timing of your voice to generate sound effects that match the dynamics of your recording.

Learn to combine your recorded voice with a text prompt, syncing the energy and timing of your voice to create sound effects.

Using Voice to sound effects (beta) can be beneficial for precisely timing sound effects to match actions, ensuring they align perfectly with intended events in the timeline. This approach helps avoid the inefficiencies of working with sound clips that are either too long or too short to match the actions and events in the video or audio timeline. 

Recording your voice is only used to guide the timing and energy of the generated sound effect. This feature does not generate speech or sung lyrics.

Original silent video

Video generated using the Text to Video feature. Prompt used: "A lion lets out a powerful roar, its gaze fixed on the horizon."

Video with recorded voice performance

Video with the voice performance recorded to control the timing and energy of the sound effect.

Video with generated sound effect

Video with the generated sound effect for the prompt “Lion roaring” with the voice performance used to control the timing and energy.

  1. On the Adobe Firefly homepage, go to the Audio module and select Voice to sound effects (beta).

  2. On the Generate Sound Effects page, select Upload media to upload a video or audio file that you want to use for composing sound effects.

    On the Generate Sound Effects page, you can upload a media file using the provided button. The page also specifies the ideal length and formats for optimal results.
    Select the Upload media option to upload a file in .mp4, .mp3, .mov, .wav, or .aac format.

  3. Position the playhead at the point in the timeline where you want to add the sound effect.

    The playhead is highlighted and positioned somewhere in the middle of the timeline. You have options to preview, add another audio track, and mute the channel.
    Navigate through the timeline and place the playhead at the exact spot where you want to add the sound effect.

  4. Under the Voice performance section, you can upload an audio file from your device or use Record voice to start recording.

    Note:

    You can record your voice for up to 30 seconds or the duration of the video, whichever is shorter, and within the allowed limit.

  5. The Recording indicator gets activated, and you can begin recording once the three-second countdown ends.

    In the Voice Performance section, there are options to either upload an audio file or record a voiceover. The recording button is prominently highlighted, allowing you to record up to 30 seconds.
    Record your voice to add a rough guide and indicate what the final generated variations should sound like.

    Use the play button in the Voice recording section to listen to the recording.

  6. In the prompt field, you'll see an indication that a voice recording has been added. Enter a text prompt describing the sound effect to be used alongside the recording to generate sound effect variations.

    For example, to get a lion roar timed to your video, enter the text prompt "lion roaring" and then record yourself roaring in sync with your video.

    Tip:

    Explore some effective text prompt writing techniques for sound effects generation.

    On the Generate Sound Effects page, there's a prompt section where you can enter a text description to generate a sound effect. The current text prompt is - 'lion roaring'.
    Add a text prompt to be used in combination with your recorded voice and generate sound effect variations.

  7. Select Generate   and select one of the generated sound effect variations to insert it at the playhead's position in the timeline.

    The available sound effect variations are displayed, and you can choose one to insert at the current playhead position on the timeline.
    Select one of the generated sound effect variations and insert it at the playhead position.

    Tip:

    To replace a sound effect in the timeline, first select the existing sound effect, then select a replacement from the generated variations.

  8. You can edit the generated clip using the available options - mute or adjust the volume. You can also download, rename, or delete it.

    The sound effect added in the timeline is selected and there are editing options to use. These editing options include mute, volume, download, and delete.
    Use the available editing options to edit the generated sound effects within the timeline directly.

    Tip:

    To trim a generated sound effects clip, drag the trim handles at either end inward to shorten it. After trimming, move the clip along the timeline to reposition it as needed.

  9. If you want to add another layer of sound effect to the timeline, select the Add Audio Track option, and follow the same process of generating it using the Prompt bar.

    Above the timeline where the video and audio are listed, the Add Audio Track option is highlighted. It also has a tooltip that mentions that you can add upto 5 audio tracks to the timeline.
    Use Add Audio Track to generate another sound effect and add it to the timeline along with the existing ones.

    Note:

    Currently, a maximum of five audio tracks are supported in addition to the uploaded media track.

  10. Use the Play   icon to preview your composition and the Download   option to download the result.