> ## Documentation Index
> Fetch the complete documentation index at: https://docs.tryhamsa.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Text to speech history: play, compare and download

> Find past text to speech takes by date and dialect, play and download the WAV audio, compare two takes with A, B and S, remix one, or delete it.

Every generation you finish is saved as a job in the **History** tab beside the script. From there you can play a take, open it, download its audio, compare two takes side by side, and start a new script from one with **Remix**.

## Find a take

Open the **History** tab (on a narrow window or a phone, the **History** button in the header). Cmd+K (Ctrl+K on Windows) also has "Open job history".

"Job history" lists your takes, newest first, with how many there are. Each row shows the start of the script, the date, the voice and its dialect, and a play button. More rows load as you scroll.

![The History tab with Filters 1, Newest, a 7 days chip and Clear filters, six takes each with voice, script start, date and dialect, the first marked Take A, and the Take A picked bar](https://mintcdn.com/hamsa-452fa79a/etUfiFCo8p_6XpY_/images/media/text-to-speech/history-tab-light.webp?fit=max\&auto=format\&n=etUfiFCo8p_6XpY_\&q=85\&s=98cdae2ad2ef29eaa09bf344e8b7f6d0)

### Filter and sort

| Control | Options | Default |
| - | - | - |
| **Filters**, **Created** | **All time**, **1 hour**, **Today**, **7 days**, **30 days** | **All time** |
| **Filters**, **Dialects** | **All dialects**, or one dialect | **All dialects** |
| **Sort** | **Newest**, **Oldest** | **Newest** |

![The Filter jobs panel under Filters, with Created set to All time and Dialects set to All dialects](https://mintcdn.com/hamsa-452fa79a/etUfiFCo8p_6XpY_/images/media/text-to-speech/history-filters-light.webp?fit=max\&auto=format\&n=etUfiFCo8p_6XpY_\&q=85\&s=14d6ed4f191b0132243da703cfd0fb37)

Active filters show as chips you can remove one by one, or all at once with **Clear filters**. **Reset** in the filter panel does the same. The count then shows the matching takes out of the total. Filters and sort reset when you reload the page.

**Refresh** reloads the list and clears the filters.

### Whose takes you see

History lists the speech you generated in the current project. Generations by other members of the project don't appear in your history, and a link to someone else's job opens "Job not found".

## Play and open a take

* Click the play button on a row to hear it. Starting another take stops the first.
* Click the row, or **View** in its menu, to open the job. The script appears read only, the **Voice settings** tab shows the voice, expressiveness and dictionaries it used, and the audio loads in the player.
* If you have unsaved text in the editor, Hamsa asks first: "Opening this job will discard your current text." **Discard and open** drops your draft.

From an open job, click **Create new** to go back to an empty script, or **Remix** to start from this job.

## Remix a take

**Remix** starts a new script with the job's script, voice and expressiveness, ready to change and generate again. The job's dictionaries are not carried over; turn on the ones you need.

If you were writing a different script, Hamsa asks before replacing it: "Text to speech has a script you were writing. Remixing this job replaces it." **Keep my script** cancels; **Replace it** goes ahead.

## Compare two takes

Compare plays two takes one after the other from the same point in the script, so you can hear the same line in both. Nothing new is generated.

**1. Pick take A**

Open a take's menu and click **Compare**. The row is marked "Take A", and a bar says "Take A picked. Pick another take from its menu to compare."

**2. Pick take B**

Click **Compare** in another take's menu. The "Compare takes" window opens with both takes, their voices, dates and scripts.

**3. Listen**

Play a take, then switch. The other take starts at the same point.

| Key | What it does |
| - | - |
| A | Play take A from the current point |
| B | Play take B from the current point |
| S | Switch to the other take at the same point |
| Space | Pause or play |

The keys work on an Arabic keyboard too. **Remove from compare** in a take's menu takes it out; **Cancel** in the bar clears the pick. If you pick a third take, the oldest pick is dropped.

![The Compare takes window with take A at 0:02 of 0:13 and take B, each with voice, date and script](https://mintcdn.com/hamsa-452fa79a/etUfiFCo8p_6XpY_/images/media/text-to-speech/compare-takes-light.webp?fit=max\&auto=format\&n=etUfiFCo8p_6XpY_\&q=85\&s=6f7988a31346f458ff59acc008f5e16f)

## Download, copy and delete

Open a take's menu (the three dots):

| Action | What it does |
| - | - |
| **Copy** | Copies the take's script ("Script copied") |
| **Download** | Downloads the audio ("Audio downloaded") |
| **Delete** | Deletes the job and its audio, after you confirm |

![A take's menu: View, Copy, Download, Compare and Delete](https://mintcdn.com/hamsa-452fa79a/etUfiFCo8p_6XpY_/images/media/text-to-speech/take-menu-light.webp?fit=max\&auto=format\&n=etUfiFCo8p_6XpY_\&q=85\&s=0b9c72a4d73b66783d5e30b2ed9a2bb5)

Audio is saved as WAV. You can also download the take that is in the player with the player's download button.

**Delete** asks first: "Delete this job? The job and its audio are deleted. You can undo this for a few seconds." After you confirm, the take leaves the list and an **Undo** stays for 6 seconds. After that, it is gone. Deleting needs permission to delete text to speech jobs (Owners, Admins and Users have it by default).

![The Delete this job? dialog: the job and its audio are deleted, this cannot be undone, with Cancel and Delete](https://mintcdn.com/hamsa-452fa79a/etUfiFCo8p_6XpY_/images/media/text-to-speech/delete-job-light.webp?fit=max\&auto=format\&n=etUfiFCo8p_6XpY_\&q=85\&s=38356d952038d80cd21d55eeb7190f33)

## Good to know

* Only finished generations are saved. A generation that fails leaves no job in History.
* A job's script and settings can't be edited. Remix it into a new script instead.
* Each take is charged once, when it is generated. Playing, comparing and downloading cost nothing.
* If the job's voice was deleted, the job still plays; its settings show "Voice not found".

## Related

> [card] [Generate speech](https://docs.tryhamsa.com/media/text-to-speech/generate-speech.md)
>
> Write a script and generate a take.

> [card] [Voice settings](https://docs.tryhamsa.com/media/text-to-speech/voice-settings.md)
>
> Change the voice or expressiveness for the next take.

> [card] [Lists and uploads](https://docs.tryhamsa.com/media/workspace/lists-and-uploads.md)
>
> The media player and its keyboard shortcuts.

> [card] [Text to speech](https://docs.tryhamsa.com/media/text-to-speech/overview.md)
>
> Voices, dialects and credits.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.