> For the full page index, fetch https://echonova.ai/llms.txt # Formats, quality and download The audio format a portal generation produces, the controls that affect its quality, the player, and how a downloaded file is named. This page tells you what file a Text to Speech generation gives you, which controls change how it sounds, what the player can do, and what the file is called when you save it. Look here before you promise a format to someone else. ## Output format | Fact | Value | | -------------------- | ------------------------------------------------------------------------------------------------------------------------- | | Format | WAV. Every generation made in the portal is a WAV file. | | Files per generation | One. | | Format options | None. The **Options** panel has no format, bit-rate or sample-rate control, and the portal does not show the sample rate. | | Duration | Shown in the player as `m:ss`, next to the elapsed time. | To get MP3 or another format, download the WAV file and convert it with a tool of your choice. ## Quality There is no format selector. Two controls in the **Options** panel change how the synthesis is run, and both are described on [Voice settings](/docs/playground/voice-settings): | Control | Values | Default | | ------------ | ------------------------------------- | ------- | | **Quality** | 16, 32, 48, 64 or 128 synthesis steps | 48 | | **Guidance** | 1.0 to 5.0 | 2.0 | Neither control changes the file format or the charge. ## The player The player opens at the bottom of the page after a generation, and again when you replay an item from **History**. It plays on its own once the file is ready. | Control | What it does | | -------------------------------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | **Play** and **Pause** | Start and stop playback. | | **Back 5 seconds** and **Forward 5 seconds** | Jump back or ahead by 5 seconds. | | Seek bar | Click or drag to move to a point in the audio. With the bar focused, the arrow keys move 5 seconds, **Space** plays and pauses, and **Home** and **End** jump to the start and the end. | | Time | The elapsed time before the bar and the total time after it, as `m:ss`. | | **Playback speed** | Cycles the playback rate through `0.75×`, `1×`, `1.25×`, `1.5×` and `2×`. It does not change the file. | | **Download** | Saves the file. See below. | | **Close** | Closes the player. It sits at the right edge of the bar. | Two notices can appear under the player controls: | Notice | Meaning | | ------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------- | | **Automatic playback was blocked. Tap Play to listen.** | The browser did not allow playback to start on its own. Click **Play**. | | **This audio could not be played in your browser.** | The browser could not play the file. Download it and play it locally, or replay it from **History** in another browser. | ## Download | Fact | Value | | -------------------------------------- | ------------------------------------------------------------------------------------------------------------------ | | Where | **Download** in the player, **Download** on a **History** item, and **Download** on the **Assets** page. | | File name from the Text to Speech page | `speech-` followed by the generation's identifier, then `.wav`, for example `speech-66e5ef3a9c1d2b4f8a7e6c01.wav`. | | File name from the Assets page | The first 40 characters of the text (or the asset title), then `.wav`. The extension follows the audio type. | | Cost | None. Downloading a generation any number of times is free. | | Who can download | Every member of the workspace, for every generation made from the portal in that workspace. | The file is saved by your browser into its usual downloads location; no save dialog is shown unless your browser always shows one. ## Worked example You generated a 12-second sentence and want to hand the file to a video editor who asked for MP3. 1. In the player, the total time reads `0:12`. Click **Download**. 2. Your downloads folder now holds `speech-66e5ef3a9c1d2b4f8a7e6c01.wav`. 3. Convert the WAV file to MP3 with your own tool and send the result. The portal charged the generation once; the download and the conversion cost nothing. ## FAQ ### Can I get MP3 from the portal? No. Every generation is WAV. Convert the downloaded file yourself. ### What is the sample rate? The portal does not show it. Read it from the downloaded file's properties in your audio tool. ### Does a second download cost credits? No. Only the generation is charged. Replaying and downloading are free. ### Why did my download fail? A notification titled **Error** appears and a new tab opens in place of the file. Close the tab, wait a moment and click **Download** again. If it keeps failing, download the same generation from **History**. ### Can I rename the file? Yes. The name is only a suggestion from the browser; rename the file after it is saved. Put the voice and the date in the name, because the identifier alone does not tell you which generation it was. ## Related - [Generate speech](/docs/playground/generate-speech): making the file in the first place - [Voice settings](/docs/playground/voice-settings): **Quality** and **Guidance** in full - [History](/docs/playground/tts-history): downloading an earlier generation - [Text to Speech voices and options](/docs/developers/tts-voices-and-options): the same WAV output over the API