> For the full page index, fetch https://echonova.ai/llms.txt # A tour of the Home page What each area of Home shows, and what every card, link and row opens. **Home** is the page every sign-in opens. Its live studio does the real work right on the page, and its other cards and links open the pages where you do the rest. Read this page to know what each area shows and where each control takes you. ## The greeting The top of the page reads **Good morning**, **Good afternoon** or **Good evening** by your clock, followed by your first name, then one line: **Write something and hear it in a voice from here. Drop a recording and read it back, or clean it.** The same line shows for every workspace, new or not. ## Generate, transcribe or clean, right here Under the greeting sits a live studio with three tabs: **Text to Speech**, **Speech to Text** and **Sound Cleaning**. Each does the real thing on Home itself, with nothing charged until you press its action: | Tab | What it does | | ------------------ | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | | **Text to Speech** | Type in the box (add **[laugh]**, **[cough]** or **[clear throat]** from **Tags**; remove one with its **−**), pick a dialect and a voice, then **Generate**. The credit estimate updates as you type. | | **Speech to Text** | Drop an audio file or record with the mic, then **Transcribe**. | | **Sound Cleaning** | Drop a recording, pick a preset (**Studio**, **Wind**, **Radio** or **Deep**), then **Clean audio**. For hand-tuned controls, use the **Sound Cleaning** page in the sidebar. | **Agents** has no card on Home; open it from **Agents** in the sidebar. ## Every voice is from somewhere A dialect strip lists every dialect that has voices, for example **Emirati**, **Saudi**, **Egyptian**, **Kuwaiti**, **Fusha**, **American** and **British**, then **Coming Soon** for the dialects announced without voices yet, such as **Qatari**, **Omani**, **Lebanese** and **Syrian**. The list comes from the voice library, so a dialect that gets voices joins it without a release. Under it, six voice cards show the voice's orb, its name and dialect, and **Play sample**; a card whose voice has no preview has the control disabled. **Open the Voice Library** opens the voice library, where every voice has a row, a page and **Use this voice**. See [Browse the voice library](/docs/playground/voice-library). If the voices cannot be loaded, the row is replaced by **We couldn't load voices.** ## Recents Labelled **Recents** and headed **What you made**, with **Open Assets** on the right, this table lists the workspace's last thirty completed jobs across Text to Speech, Speech to Text and Sound Cleaning, newest first, in four columns: **Text or file**; **Voice** and dialect, for a Text to Speech row; **When**; and **Length**. **Open Assets** opens the **Assets** page, where every item can be played, downloaded and started over; see [Assets and history](/docs/playground/assets-and-history). Selecting a **Text to Speech** row opens the Text to Speech page with that generation in the player and its text in the writing area; it plays, and nothing is charged. Selecting a **Speech to Text** row opens the Speech to Text page; the transcript is not opened there, read it on **Assets**. Selecting a **Sound Cleaning** row opens the Sound Cleaning page. Before anything has been made, the table shows **Nothing here yet.** with a line explaining that the first generation, transcript or cleanup lands there. If the list cannot be loaded, it reads **We couldn't load your recents.** ## On the way A single card, headed **Coming soon**, reads **On the way.** and **Five more tools, in this order.**, followed by **EchoReader**, **Dubbing**, **Sound Effects**, **Audiobooks** and **Arab Voices**. Nothing in the card opens a page. ## Around the page The sidebar and the header are the same on every page. On the sidebar: | Group | Entries | What opens | | -------------- | ---------------------------------------------------------------------- | ------------------------------------------------------------------------------ | | Top | **Home**, **Voices** | This page; the voice library | | **Build** | **Agents** | The agents list | | **Playground** | **Text to Speech**, **Speech to Text**, **Sound Cleaning**, **Assets** | Each service's page, and the workspace's record of what they made | | **Developers** | **API Key**, **Documentation** | Developer Settings, where API keys and webhooks live; these docs, in a new tab | | **More Tools** | Names marked **Coming Soon** | Nothing | Under the groups sits the plan card: the plan's name, the credits used out of the plan's total with the label **Used**, a thin meter, a **Low** flag when 15% or less of the plan's credits remain, and **Upgrade Your Plan**, which opens the Subscription page. What the numbers count is on [Track your credit usage](/docs/account/usage), and where the credits come from on [Credits, top-ups and rollover](/docs/account/credits-top-ups-rollover). The header carries the page title, the notification bell and the user menu. The bell is described on [Read your notifications](/docs/account/notifications). The user menu shows **Credits remaining** and opens **Account settings** and **Subscription & billing**, sets **Appearance** (**Light**, **Dark**, **System**) and **Language**, and ends with **Log out**. ## Worked example On Sara's first visit, **Home** greets her and the **Text to Speech** tab is already open with a sample line typed in for the **Emirati** dialect. She picks a voice, presses **Generate**, and hears it play back with nothing charged yet; **Recents** is empty, reading **Nothing here yet.** She types her own sentence, picks a voice and presses **Generate** again; this one is charged, and it now appears at the top of **Recents** with her voice, its dialect and how long ago she made it. She wants to hear it again, so she selects the row: the Text to Speech page opens with her sentence in the player and plays it, for free. ## FAQ ### Why is Recents empty? Nothing has been generated, transcribed or cleaned in the workspace yet. A row appears the moment the first job finishes. ### Does Home show Sound Cleaning jobs? Yes. **Recents** covers Text to Speech, Speech to Text and Sound Cleaning together, newest first. See [Clean a recording](/docs/playground/clean-a-recording). ### Can I replay a generation from Home for free? Yes. Select its row in **Recents**: the Text to Speech page opens with it in the player. **Open Assets** and play from **Assets** is free too. ### Why does the voice row not show the voice I saved? The row is the six-voice spread the design ships with, and it does not follow your saved voices. Open **Voices** and use **My Saved Voices**; see [Filters, previews and favourites](/docs/playground/filters-previews-favourites). ### Where is my workspace's name? Not on **Home**, and not on the other portal pages; see [Workspaces](/docs/account/workspaces#where-do-i-see-my-workspace-s-name). ## Related - [Quickstart: your first speech file](/docs/get-started/quickstart-first-speech): what to press first on this page - [Quickstart: your first agent](/docs/get-started/quickstart-first-agent): starting from **Agents** in the sidebar, carried through to a connected channel - [Browse the voice library](/docs/playground/voice-library): where **Open the Voice Library** on the voices row takes you - [Assets and history](/docs/playground/assets-and-history): where **Open Assets** on Recents takes you - [Track your credit usage](/docs/account/usage): what the plan card and **Credits remaining** count, and the ledger behind them