diff --git a/marker/v2/builders/structure.py b/marker/v2/builders/structure.py
index 9ff6c25..f73d7be 100644
--- a/marker/v2/builders/structure.py
+++ b/marker/v2/builders/structure.py
@@ -1,8 +1,9 @@
-from typing import List, Optional
+from typing import Optional
 
 from pydantic import BaseModel
 
 from marker.v2.builders import BaseBuilder
+from marker.v2.schema import BlockTypes
 from marker.v2.schema.document import Document
 from marker.v2.schema.groups import GROUP_BLOCK_REGISTRY, ListGroup
 from marker.v2.schema.groups.page import PageGroup
@@ -22,7 +23,7 @@ def __call__(self, document: Document):
     def group_caption_blocks(self, page: PageGroup):
         for i, block_id in enumerate(page.structure):
             block = page.get_block(block_id)
-            if block.block_type not in ["Table", "Figure", "Picture"]:
+            if block.block_type not in [BlockTypes.Table, BlockTypes.Figure, BlockTypes.Picture]:
                 continue
 
             block_structure = [block_id]
@@ -30,7 +31,7 @@ def group_caption_blocks(self, page: PageGroup):
             for j, prev_block_id in enumerate(page.structure[:i][::-1]):
                 prev_block = page.get_block(prev_block_id)
                 if all([
-                    prev_block.block_type in ["Caption", "Footnote"],
+                    prev_block.block_type in [BlockTypes.Caption, BlockTypes.Footnote],
                     prev_block.polygon.minimum_gap(block.polygon) < self.gap_threshold
                 ]):
                     block_structure.insert(0, prev_block_id)
@@ -41,7 +42,7 @@ def group_caption_blocks(self, page: PageGroup):
             for j, next_block_id in enumerate(page.structure[i + 1:]):
                 next_block = page.get_block(next_block_id)
                 if all([
-                    next_block.block_type in ["Caption", "Footnote"],
+                    next_block.block_type in [BlockTypes.Caption, BlockTypes.Footnote],
                     next_block.polygon.minimum_gap(block.polygon) < self.gap_threshold
                 ]):
                     block_structure.append(next_block_id)
@@ -63,7 +64,7 @@ def group_caption_blocks(self, page: PageGroup):
     def group_lists(self, page: PageGroup):
         for i, block_id in enumerate(page.structure):
             block = page.get_block(block_id)
-            if block.block_type not in ["ListItem"]:
+            if block.block_type not in [BlockTypes.ListItem]:
                 continue
             block_structure = [block_id]
             selected_polygons = [block.polygon]
@@ -71,7 +72,7 @@ def group_lists(self, page: PageGroup):
             for j, next_block_id in enumerate(page.structure[i + 1:]):
                 next_block = page.get_block(next_block_id)
                 if all([
-                    next_block.block_type == "ListItem",
+                    next_block.block_type == BlockTypes.ListItem,
                     next_block.polygon.minimum_gap(block.polygon) < self.gap_threshold
                 ]):
                     block_structure.append(next_block_id)
diff --git a/marker/v2/converters/pdf.py b/marker/v2/converters/pdf.py
index e4774c0..27ea2a4 100644
--- a/marker/v2/converters/pdf.py
+++ b/marker/v2/converters/pdf.py
@@ -13,6 +13,8 @@
 from marker.v2.providers.pdf import PdfProvider
 from marker.v2.models import setup_layout_model, setup_texify_model, setup_recognition_model, setup_table_rec_model, \
     setup_detection_model
+from marker.v2.renderers.line import LineRenderer
+from marker.v2.renderers.span import SpanRenderer
 
 
 class PdfConverter(BaseConverter):
@@ -39,7 +41,8 @@ def __call__(self, filepath: str, page_range: List[int] | None = None):
         #table_processor = TableProcessor(self.detection_model, self.recognition_model, self.table_rec_model)
         #table_processor(document)
 
-        rendered = document.render()
+        renderer_lst = [SpanRenderer(), LineRenderer()]
+        rendered = document.render(renderer_lst)
         return rendered
 
 
diff --git a/marker/v2/processors/table.py b/marker/v2/processors/table.py
index e2b1b08..fa62df0 100644
--- a/marker/v2/processors/table.py
+++ b/marker/v2/processors/table.py
@@ -7,11 +7,12 @@
 
 from marker.settings import settings
 from marker.v2.processors import BaseProcessor
+from marker.v2.schema import BlockTypes
 from marker.v2.schema.document import Document
 
 
 class TableProcessor(BaseProcessor):
-    block_type = "Table"
+    block_type = BlockTypes.Table
     detect_boxes = False
     detector_batch_size = None
     table_rec_batch_size = None
diff --git a/marker/v2/providers/pdf.py b/marker/v2/providers/pdf.py
index 4c9b391..782d6ea 100644
--- a/marker/v2/providers/pdf.py
+++ b/marker/v2/providers/pdf.py
@@ -1,5 +1,5 @@
 import functools
-from typing import Dict, List, Tuple
+from typing import Dict, List, Tuple, Set
 from typing import Dict, List, Optional
 
 import pypdfium2 as pdfium
@@ -34,7 +34,7 @@ def __len__(self) -> int:
     def __del__(self):
         self.doc.close()
 
-    def font_flags_to_format(self, flags: int) -> List[str]:
+    def font_flags_to_format(self, flags: int) -> Set[str]:
         flag_map = {
             1: "FixedPitch",
             2: "Serif",
@@ -69,7 +69,15 @@ def font_flags_to_format(self, flags: int) -> List[str]:
                 formats.add("bold")
             if set_flags & {"FixedPitch", "Serif", "Script", "Nonsymbolic", "AllCap", "SmallCap", "UseExternAttr"}:
                 formats.add("plain")
-        return list(formats)
+        return formats
+
+    def font_names_to_format(self, font_name: str) -> Set[str]:
+        formats = set()
+        if "bold" in font_name.lower():
+            formats.add("bold")
+        if "ital" in font_name.lower():
+            formats.add("italic")
+        return formats
 
     def setup(self):
         self.doc = pdfium.PdfDocument(self.filepath)
@@ -90,6 +98,7 @@ def setup(self):
                     for span in line["spans"]:
                         if not span["text"].strip():
                             continue
+                        font_formats = self.font_flags_to_format(span["font"]["flags"]).union(self.font_names_to_format(span["font"]["name"]))
                         spans.append(
                             Span(
                                 polygon=PolygonBox.from_bbox(span["bbox"]),
@@ -99,7 +108,7 @@ def setup(self):
                                 font_size=span["font"]["size"],
                                 minimum_position=span["char_start_idx"],
                                 maximum_position=span["char_end_idx"],
-                                formats=self.font_flags_to_format(span["font"]["flags"]),
+                                formats=list(font_formats),
                                 page_id=page_id,
                             )
                         )
diff --git a/marker/v2/renderers/__init__.py b/marker/v2/renderers/__init__.py
index 8806b78..00616c9 100644
--- a/marker/v2/renderers/__init__.py
+++ b/marker/v2/renderers/__init__.py
@@ -1,8 +1,14 @@
+from enum import Enum
 from typing import Optional
 
 from pydantic import BaseModel
 
 
+class RenderFormat(str, Enum):
+    json = "json"
+    markdown = "markdown"
+
+
 class BaseRenderer:
     block_type: str | None = None
 
diff --git a/marker/v2/renderers/line.py b/marker/v2/renderers/line.py
new file mode 100644
index 0000000..fb22227
--- /dev/null
+++ b/marker/v2/renderers/line.py
@@ -0,0 +1,41 @@
+import re
+from typing import List, Optional
+
+from marker.v2.renderers import BaseRenderer
+from marker.v2.schema import BlockTypes
+from marker.v2.schema.text import Span
+
+
+def surround_text(s, char_to_insert):
+    leading_whitespace = re.match(r'^(\s*)', s).group(1)
+    trailing_whitespace = re.search(r'(\s*)$', s).group(1)
+    stripped_string = s.strip()
+    modified_string = char_to_insert + stripped_string + char_to_insert
+    final_string = leading_whitespace + modified_string + trailing_whitespace
+    return final_string
+
+
+class LineRenderer(BaseRenderer):
+    block_type = BlockTypes.Line
+
+    def __call__(self, document, block, children: Optional[List[Span]] = None):
+        text = ""
+        for i, child in enumerate(children):
+            next_span = None
+            next_idx = i + 1
+            while len(children) > next_idx:
+                next_span = children[next_idx]
+                next_idx += 1
+                if len(next_span.text.strip()) > 0:
+                    break
+            span_text = child.rendered
+
+            # Don't bold or italicize very short sequences
+            # Avoid bolding first and last sequence so lines can be joined properly
+            if len(span_text) > 3 and 0 < i < len(children) - 1:
+                if child.italic and (not next_span or not next_span.italic):
+                    span_text = surround_text(span_text, "*")
+                elif child.bold and (not next_span or not next_span.bold):
+                    span_text = surround_text(span_text, "**")
+            text += span_text
+        return text
\ No newline at end of file
diff --git a/marker/v2/renderers/span.py b/marker/v2/renderers/span.py
new file mode 100644
index 0000000..ada64ff
--- /dev/null
+++ b/marker/v2/renderers/span.py
@@ -0,0 +1,9 @@
+from marker.v2.renderers import BaseRenderer
+from marker.v2.schema import BlockTypes
+
+
+class SpanRenderer(BaseRenderer):
+    block_type = BlockTypes.Span
+
+    def __call__(self, document, block, children=None):
+        return block.text
\ No newline at end of file
diff --git a/marker/v2/renderers/util.py b/marker/v2/renderers/util.py
index 335cf45..e9a301a 100644
--- a/marker/v2/renderers/util.py
+++ b/marker/v2/renderers/util.py
@@ -5,4 +5,4 @@ def renderer_for_block(block, renderer_list: list):
         if renderer.block_type == block.block_type:
             return renderer
 
-    return DefaultRenderer
+    return DefaultRenderer()
diff --git a/marker/v2/schema/__init__.py b/marker/v2/schema/__init__.py
index 8b13789..ffb8cb0 100644
--- a/marker/v2/schema/__init__.py
+++ b/marker/v2/schema/__init__.py
@@ -1 +1,18 @@
+from marker.v2.schema.blocks import Block, LAYOUT_BLOCK_REGISTRY
+from marker.v2.schema.groups import GROUP_BLOCK_REGISTRY
+from marker.v2.schema.text import TEXT_BLOCK_REGISTRY
 
+
+class _BlockTypes:
+    def __init__(self):
+        pass
+
+    def add(self, registry: dict[str, Block]):
+        for k, v in registry.items():
+            setattr(self, k, k)
+
+
+BlockTypes = _BlockTypes()
+BlockTypes.add(GROUP_BLOCK_REGISTRY)
+BlockTypes.add(TEXT_BLOCK_REGISTRY)
+BlockTypes.add(LAYOUT_BLOCK_REGISTRY)
diff --git a/marker/v2/schema/blocks/base.py b/marker/v2/schema/blocks/base.py
index d7bed64..5331a7a 100644
--- a/marker/v2/schema/blocks/base.py
+++ b/marker/v2/schema/blocks/base.py
@@ -91,6 +91,5 @@ def render(self, document, renderer_list: list):
                     block.render(document, renderer_list)
                 child_blocks.append(block)
 
-        renderer_cls = renderer_for_block(self, renderer_list)
-        renderer = renderer_cls()
+        renderer = renderer_for_block(self, renderer_list)
         self.rendered = renderer(document, self, child_blocks)
diff --git a/marker/v2/schema/blocks/inlinemath.py b/marker/v2/schema/blocks/inlinemath.py
index 4cc7638..2108046 100644
--- a/marker/v2/schema/blocks/inlinemath.py
+++ b/marker/v2/schema/blocks/inlinemath.py
@@ -2,4 +2,4 @@
 
 
 class InlineMath(Block):
-    block_type: str = "Text-inline-math"
+    block_type: str = "TextInlineMath"
diff --git a/marker/v2/schema/blocks/listitem.py b/marker/v2/schema/blocks/listitem.py
index 0e3f67e..3f13849 100644
--- a/marker/v2/schema/blocks/listitem.py
+++ b/marker/v2/schema/blocks/listitem.py
@@ -2,4 +2,4 @@
 
 
 class ListItem(Block):
-    block_type: str = "List-item"
+    block_type: str = "ListItem"
diff --git a/marker/v2/schema/blocks/pagefooter.py b/marker/v2/schema/blocks/pagefooter.py
index a676a98..329885c 100644
--- a/marker/v2/schema/blocks/pagefooter.py
+++ b/marker/v2/schema/blocks/pagefooter.py
@@ -2,4 +2,4 @@
 
 
 class PageFooter(Block):
-    block_type: str = "Page-footer"
+    block_type: str = "PageFooter"
diff --git a/marker/v2/schema/blocks/pageheader.py b/marker/v2/schema/blocks/pageheader.py
index e48dc21..3215073 100644
--- a/marker/v2/schema/blocks/pageheader.py
+++ b/marker/v2/schema/blocks/pageheader.py
@@ -2,4 +2,4 @@
 
 
 class PageHeader(Block):
-    block_type: str = "Page-header"
+    block_type: str = "PageHeader"
diff --git a/marker/v2/schema/blocks/sectionheader.py b/marker/v2/schema/blocks/sectionheader.py
index 01f0b74..046ddbc 100644
--- a/marker/v2/schema/blocks/sectionheader.py
+++ b/marker/v2/schema/blocks/sectionheader.py
@@ -2,4 +2,4 @@
 
 
 class SectionHeader(Block):
-    block_type: str = "Section-header"
+    block_type: str = "SectionHeader"
diff --git a/marker/v2/schema/blocks/toc.py b/marker/v2/schema/blocks/toc.py
index 1179633..8bfeee1 100644
--- a/marker/v2/schema/blocks/toc.py
+++ b/marker/v2/schema/blocks/toc.py
@@ -2,4 +2,4 @@
 
 
 class TableOfContents(Block):
-    block_type: str = "Table-of-contents"
+    block_type: str = "TableOfContents"
diff --git a/marker/v2/schema/document.py b/marker/v2/schema/document.py
index 0f3dbc9..ef03764 100644
--- a/marker/v2/schema/document.py
+++ b/marker/v2/schema/document.py
@@ -12,6 +12,7 @@
 class Document(BaseModel):
     filepath: str
     pages: List[PageGroup]
+    block_type: str = "Document"
 
     def get_block(self, block_id: BlockId):
         for page in self.pages:
@@ -27,6 +28,5 @@ def render(self, renderer_lst: list | None = None):
         for page in self.pages:
             page.render(self, renderer_lst)
 
-        doc_renderer_cls = renderer_for_block(self, renderer_lst)
-        doc_renderer = doc_renderer_cls()
+        doc_renderer = renderer_for_block(self, renderer_lst)
         return doc_renderer(self, self, self.pages)
diff --git a/marker/v2/schema/text/__init__.py b/marker/v2/schema/text/__init__.py
new file mode 100644
index 0000000..1a01f22
--- /dev/null
+++ b/marker/v2/schema/text/__init__.py
@@ -0,0 +1,9 @@
+from marker.v2.schema.text.line import Line
+from marker.v2.schema.text.span import Span
+
+
+TEXT_BLOCK_REGISTRY = {
+    "Line": Line,
+    "Span": Span,
+}
+
diff --git a/marker/v2/schema/text/line.py b/marker/v2/schema/text/line.py
index a9234d7..545930c 100644
--- a/marker/v2/schema/text/line.py
+++ b/marker/v2/schema/text/line.py
@@ -6,3 +6,6 @@
 
 class Line(Block):
     block_type: str = "Line"
+
+    def is_continuation(self, other):
+        pass
diff --git a/marker/v2/schema/text/span.py b/marker/v2/schema/text/span.py
index a0af73b..cce09b4 100644
--- a/marker/v2/schema/text/span.py
+++ b/marker/v2/schema/text/span.py
@@ -13,3 +13,11 @@ class Span(Block):
     minimum_position: int
     maximum_position: int
     formats: List[Literal['plain', 'math', 'chemical', 'bold', 'italic']]
+
+    @property
+    def bold(self):
+        return 'bold' in self.formats
+
+    @property
+    def italic(self):
+        return 'italic' in self.formats